1const __vite__mapDeps=(i,m=__vite__mapDeps,d=(m.f||(m.f=["assets/WorkforceHome-BOB3kghP.js","assets/vendor-motion-VOv-vRjq.js","assets/vendor-react-CgwpgauT.js","assets/PlanExternalCosts-C5_V9m8H.js","assets/vendor-i18n-T2j15knT.js","assets/info-WbLZzUUM.js","assets/check-CVdn6-qb.js","assets/users-BkG1F9p3.js","assets/building-2-PaHduIXR.js","assets/crown-iXstI4Zx.js","assets/employees-CDLFQJh4.js","assets/connectors-Ck33qzzU.js","assets/headphones-C4MAu-E-.js","assets/megaphone-WD_5lyyv.js","assets/wallet-BY8Rhccp.js","assets/microscope-BsT2nLK2.js","assets/scale-DnSTj_3S.js","assets/cpu-DJAK3PQT.js","assets/network-Cy-xWgPy.js","assets/bot-C0eO1kvV.js","assets/clock-Bg4kWMOO.js","assets/trending-up-Dbs1zDd-.js","assets/shield-check-D9lBGIcQ.js","assets/code-xml-DiUjW4G-.js","assets/sparkles-Dr5zAGnQ.js","assets/search-CCiiTnL5.js","assets/database-Dw9LjYDX.js","assets/user-plus-8MUJ2ESg.js","assets/vendor-ui-kSdzIpai.js","assets/vendor-charts-BeJd8Fxm.js","assets/vendor-supabase-DYXPcfum.js","assets/MasterOrchestrator-CcrMYKer.js","assets/badge-NWGzI7tl.js","assets/card-E5s5bunD.js","assets/life-buoy-5z-O2x80.js","assets/shopping-bag-BGsMOUHT.js","assets/rocket-DqpZ6bts.js","assets/workflow-BpdMXb20.js","assets/circle-check-BtM-dNzi.js","assets/circle-Bo6HnW11.js","assets/globe-B3hV3p0z.js","assets/server-BRQk_4qi.js","assets/credit-card-CizH29wA.js","assets/chart-column-CyD17amq.js","assets/cloud-Jo2L4mcx.js","assets/book-open-BLy67MAN.js","assets/useUserSubscription-DDtNbVvk.js","assets/useQuery-B3l7Wx7i.js","assets/AuthContext-dDkX-q7D.js","assets/wand-sparkles-CAPKUtNV.js","assets/index-DbshHtNu.js","assets/loader-circle-vXUzU19Z.js","assets/send-8D-sWKT9.js","assets/gauge-IGjQ15vx.js","assets/CompanyBuilder-UNPHpAgf.js","assets/textarea-AhH_qBdT.js","assets/progress-DXHAMKUJ.js","assets/tabs-C1__JsOw.js","assets/shopping-cart-CnTg1wMD.js","assets/refresh-ccw-CmZXvAtB.js","assets/file-text-D9gUE6of.js","assets/languages-U0V2J0fj.js","assets/circle-x-BFRNhvLe.js","assets/dollar-sign-CtUdpZ9f.js","assets/git-branch-CdROrFje.js","assets/layers-C7awvk1E.js","assets/triangle-alert-DhIXaYD8.js","assets/briefcase-DUKHYrH-.js","assets/target-CfEcMtvq.js","assets/calendar-days-D4WNVlU4.js","assets/layout-dashboard-B2fJsuBO.js","assets/file-chart-column-increasing-B5ce8s6l.js","assets/plug-DIWUVNkK.js","assets/pen-line-BlIp67s3.js","assets/landmark-Dzz8IWWM.js","assets/list-checks--msyImKP.js","assets/earth-C6vN-dEs.js","assets/link-2-DOczT8Lb.js","assets/slider-Dl07q-6h.js","assets/index-BdQq_4o_.js","assets/index-N0Md8JAg.js","assets/shield-alert-UAU5e5aR.js","assets/lightbulb-Dl1_0SbY.js","assets/palette-Bpzh2QSJ.js","assets/ReadyMadeCompanies-B_pVX5Iz.js","assets/PublicSiteRenderer-BA3zcR5x.js","assets/SuperAdminOS-YaD0Bxzy.js","assets/timer-C8LgDWH2.js","assets/flame-CduECcXA.js","assets/heart-pulse-xOZxW4b9.js","assets/useDynamicTranslate-B3lKRJel.js","assets/copy-YaCwpMxe.js","assets/circle-play-P5SfrlPS.js","assets/RC2InnovationCenter-DBHC5uB8.js","assets/arrow-left-Bjjsfe-_.js","assets/chevron-right-K00PPZ33.js","assets/IntegrationCenter-DFRtLC6m.js","assets/MasterCommandCenter-CHDdbs1Y.js","assets/ApprovalGovernance-I-agcrzb.js","assets/trash-2-DU7MI4dm.js","assets/AuthShell-vtmnbNYL.js","assets/CookieConsent-BxH8aywZ.js","assets/settings-Dh_tQYmw.js","assets/AmyHomeGreeting-Bn8UMNLL.js","assets/MobileConversionLayer-DTDRhKeu.js","assets/ExitIntentPopup-UWTkQUll.js","assets/gift-BAFH-gg7.js","assets/FoundersPopup-4qN-7rLg.js","assets/founderTracking-CpVTgI6Z.js","assets/PlatformCopilot-BXMQLwi5.js","assets/FoundersOffer-DVf4g8hu.js","assets/label-DIh-xDUi.js","assets/useConfetti-rRY49OaX.js","assets/circle-alert-DFxkMQ2g.js","assets/HowItWorks-CZvM4I85.js","assets/BookDemo-IWnc70zQ.js","assets/Navbar-BWnMQenR.js","assets/Footer-B-jlUJcA.js","assets/types-CeNoQn1N.js","assets/Login-BG0C7Z04.js","assets/index-D-PyGsyi.js","assets/eye-off-B3bFnC5L.js","assets/eye-DGewSVWL.js","assets/Signup-DHjDxGPM.js","assets/refresh-cw-DxX1QLTW.js","assets/OAuthConsent-O7JmmFIu.js","assets/Admin-BBXpUMsr.js","assets/skeleton-Bbl0cxSq.js","assets/select-DT0yeclq.js","assets/table-yGTWkj1L.js","assets/alert-dialog-CfLWVbNS.js","assets/useMutation-BOfKT21k.js","assets/download-oZV2XGwg.js","assets/chevron-left-DhwxurBn.js","assets/formatDistanceToNow-BKI0csu8.js","assets/trending-down-C9cTJPkb.js","assets/gamepad-2-fOS_xV5h.js","assets/calendar-BgmDW69C.js","assets/star-CdnOpROH.js","assets/file-spreadsheet-CFxfzHC-.js","assets/upload-CIftz8xf.js","assets/message-square-Cu-D2q3f.js","assets/key-D4Xp46kT.js","assets/play-BJtoImz6.js","assets/settings-2-BlsbXMOS.js","assets/history-8405_pOh.js","assets/percent-Dj3a3EFD.js","assets/dropdown-menu-DhtHvy_r.js","assets/radar-Q8--p5tp.js","assets/IAOChat-CVmUhkJP.js","assets/useIAOConversations-B-4Lcs3Z.js","assets/scroll-area-Divr5Axe.js","assets/plus-S6NW74Le.js","assets/use-mobile-DSCFWEh8.js","assets/webhook-CFmmeDak.js","assets/log-out-ae1FGF6_.js","assets/Dashboard-DNWToVhZ.js","assets/code-BxD_3vjh.js","assets/useCurrency-DBPdF8lT.js","assets/shield-off-D1yq_fmU.js","assets/terminal-Bo0MOTnz.js","assets/RoleSwitcher-CF4wmBzd.js","assets/PaygDashboard-B-J
1nwr3w.js","assets/bell-D_Dy5gWV.js","assets/Privacy-DNecwcrS.js","assets/BackHome-CEtYm_V-.js","assets/Cgu-ie82Y6d5.js","assets/PaymentSuccess-DIHgWumO.js","assets/EnterpriseCheckout-CDSsT2oa.js","assets/separator-DY4QBDEG.js","assets/NotFound-Dp7yXHa2.js","assets/Blog-Cby938Pg.js","assets/AnimatedSection-C7MCm51h.js","assets/BlogTags-CdJJl0kp.js","assets/seoUtils-C67Nn7Ar.js","assets/CheapInference-B0qB4dl0.js","assets/SocialShare-2-BdOEB7.js","assets/twitter-BBSMOGdB.js","assets/OpenAIAlternative-D8l7kd5v.js","assets/trophy-Bv42_hZp.js","assets/EconomicLLM-DWOtE3_N.js","assets/calculator-NYRejV6O.js","assets/DynamicBlogPost-DETivjn8.js","assets/purify.es-C5KSVp3G.js","assets/AzureVsInfernex-BT5rDawy.js","assets/GoogleCloudAiCosts-Dj3uHtwB.js","assets/InferenceCostPerformance-BiuShDe9.js","assets/Azure70VsInfernex95-D4jnmQxB.js","assets/SemanticCacheLLM-Cx4_tc_N.js","assets/QuantumInfernex-CZ182VjT.js","assets/CompareOpenAI-CF7cP_i4.js","assets/CompareAWS-BigCrZ-J.js","assets/CompareGoogle-Dp69bhDM.js","assets/CompareGroq-C3T-S9Wa.js","assets/CompareTogether-CkAwYDKU.js","assets/CompareAnyscale-PhJPFZiE.js","assets/CompareNvidia-7AfoDFkj.js","assets/CompareHuggingFace-Ctkk7YnS.js","assets/Status-BdDuOq_N.js","assets/Changelog-Bp9pHtJm.js","assets/wrench-BSuIeNJM.js","assets/Playground-uBdTiCnj.js","assets/Calculator-BS53-_Ij.js","assets/checkbox-ChK2FjDo.js","assets/leaf-DiYlqvPH.js","assets/atom-DLaAIEtM.js","assets/Onboarding-CTj7SuXy.js","assets/house-XW-N074O.js","assets/ResetPassword-GeBd3TM8.js","assets/EmailPreview-Dm79NUBP.js","assets/WhiteLabel-sN-pSyFl.js","assets/ServiceAdvisorChat-C3tlCIfU.js","assets/Unsubscribe-CCN_WSoQ.js","assets/Partners-A8JS8KLL.js","assets/BackToTop-9za2KRXd.js","assets/PartnerProgram-CiFU_oJw.js","assets/Quantum-COjH-3FL.js","assets/EngineReminder-Br1DH9Ub.js","assets/AnimatedCounter-Bytaz9AF.js","assets/GlowingCard-BbAGuIRa.js","assets/ServiceExplainerBlock-CtpQ57Ib.js","assets/quantum-BgvOHYdr.js","assets/brain-circuit-DlyZTAVE.js","assets/file-check-DGBm0ofJ.js","assets/chart-no-axes-column-increasing-LxwtLvcI.js","assets/QuantumVsQuant-Cmd_FThO.js","assets/QuantumAds-c437Uzdb.js","assets/switch-BFuaxS93.js","assets/QuantumDashboard-D7dpgqEC.js","assets/QuantumDocs-BThuSqa0.js","assets/QuantumOnboarding-B_B4DQKi.js","assets/ReduceAICosts-CLFrLZDo.js","assets/AIInferencePricing-hhxagn1k.js","assets/WorkforcePricing-AJzYFnWb.js","assets/minus-Dy2M9WOz.js","assets/WorkforcePlanDetail-BwyTg1bp.js","assets/HireQuantAlternative-DzZwYjTH.js","assets/EnterpriseSolutions-DwxnJKyF.js","assets/QuantumAcceleratedInference-BXlICopR.js","assets/ServerlessAIInference-DVLV4wS-.js","assets/OpenAIMigrationGuide-Q2ABqRk5.js","assets/I18nAudit-DYvEui24.js","assets/SiteAudit-D7wXTf5P.js","assets/GscDashboard-B9ctj54f.js","assets/mouse-pointer-click-B2LZ0siN.js","assets/WorkforceLaunch-B83mZ01a.js","assets/LeadsAnalytics-WUJRh-o3.js","assets/TranslationGenerator-d33v4Zug.js","assets/Photonique-Cq16dDFI.js","assets/HardwareVsApiKeys-zM1I1iEo.js","assets/TSUPBit-Cu3uRXJV.js","assets/UltimateSupremacy-YrssIwoN.js","assets/CompareAnthropic-3tVCGkVg.js","assets/CompareMistral-COcVpi5T.js","assets/SwitchFromOpenAI-CkDFx3Ik.js","assets/Models-BuJ2MFCI.js","assets/GEO-CCn5FnzL.js","assets/GeoDashboard-bubhzVMM.js","assets/MiroFish-CrTk1eZu.js","assets/accordion-Ch7WgIT6.js","assets/MiroFishInference-hAYTZLi0.js","assets/MiroFishSimulation-ZHe87aPg.js","assets/MiroFishFAQ-BWDypKSx.js","assets/MiroFishDocs-D1X-eRr_.js","assets/IAO-AHRPhwlm.js","assets/IAOChat-DPFZkh-U.js","assets/IAODemo-BWT0Ywht.js","assets/Platform-xVIMJNoz.js","assets/quote-CuuToJlO.js","assets/WidgetIAOChat-CsYM5xlz.js","assets/IAOWidgetInstall-CncRLTc6.js","assets/AuditResult-DPxhqNHZ.js","assets/AuditDashboard-B7GzuI8c.js","assets/QuadFusionAGI-BQ0vh3s2.js","assets/Supremacy-DePqTTw-.js","assets/PartnerPage-BEwm8aJM.js","assets/TrustCenter-C1r8rJpG.js","assets/Unsubscribe-CVh-kklx.js","assets/Themis-Dz1vj9lh.js","assets/stamp-BrRAWzkI.js","assets/bitcoin-DelPu63P.js","assets/ThemisDashboard-Cg5JtjPq.js","assets/ThemisVerify-BG_PwPoM.js","assets/BenchmarkVerify-CY0wNvzf.js","assets/Switch-D1ezqIr9.js","assets/VsOpenAILive-SVBCEvpU.js","assets/Affiliates-CqSdlszC.js","assets/Sdk-DUII8XDd.js","assets/package-jok2YHbn.js","assets/Developers-X0pxOUor.js","assets/Testimonials-CV95L0yx.js","assets/Support-4P5X6JVt.js","assets/Founder-CoS0VsAd.js","assets/EnterpriseLayout-DGwMxaUV.js","assets/banknote-ChBE_Hr5.js","assets/KpiCard-BIjZ6G8h.js","assets/Investor-DS-Yn7lB.js","assets/MiroFishNetworkCard-CCR2f3rU.js","assets/Documents-DGj_yi6o.js","assets/CRM-CNBrgkKN.js","assets/PageHeader-ta1BLbk9.js","assets/DataTable-BVKhTEp-.js","assets/LeadScoreBadge-DPDdnYxK.js","assets/EnterpriseAccounts-CYgLAGdD.js","assets/PartnerCenter-BFAbeCNU.js","assets/key-round-BSK3b86l.js","assets/MarketplaceCenter-CpVHnHDT.js","assets/file-down-CJrT3StB.js","assets/FinOps-B3_6avt3.js","assets/BizDev-C349iLg4.js","assets/DeveloperConsole-BndKbfMe.js","assets/DataRoom-DFRRwqYf.js","assets/PitchDeck-C8YQwpOa.js","assets/AiVerificationBadge-lFuiIv1i.js","assets/jspdf.es.min-ComJkagf.js","assets/DueDiligence-DZTiFAEg.js","assets/PredictiveAnalytics-CWdnnxad.js","assets/Copilot-DMCf8rAS.js","assets/AiSupervisor-
1DMMVqVxC.js","assets/RoiEngine-BDWhuo3K.js","assets/AutomationCenter-De59C7hN.js","assets/Directories-Bj0_cXiB.js","assets/PartnershipRoadmap-DDEkYiPx.js","assets/KnowledgeCenter-DbjzGG_2.js","assets/StubPage-cxiC0Fok.js","assets/SecurityCenter-DHH_0sXJ.js","assets/GlobalOptimizer-CDp4Fvga.js","assets/useMiroFish-BXRAYABU.js","assets/ExecutiveInsights-DBYz0A1D.js","assets/UserAnalytics-jsJBM76b.js","assets/WarRoomYC-BV1mnJbJ.js","assets/AiWorkforce-DYv148JU.js","assets/AiWorkforceOnboarding-F0DLQ7Kn.js","assets/AiWorkforceChat-CbVUZ21P.js","assets/WorkforceDashboard-vi1DxFfr.js","assets/HomeLink-vr2HZzVB.js","assets/WorkforceConnectors-CPrnlNsG.js","assets/WorkforceRouter-BEhN72-I.js","assets/WorkforceSafety-DX7uHaqM.js","assets/GlobalIntelligenceCenter-uk-Zke9p.js","assets/PartnerIntelligenceCenter-CDOx3I2q.js","assets/IntelligenceAlertsBar-DblNHVQZ.js","assets/FinOpsIntelligenceCenter-CwUu_Y46.js","assets/ComplianceCenter-Cen3h48p.js","assets/ExecutiveCommandCenter-CJdSM2dy.js","assets/QualityGate-BXj2QXOs.js","assets/KnowledgeGraph-BF0HQQ6D.js","assets/PartnerPortal-XJ0f_wF_.js","assets/CustomerPortal-C7IOr_Iv.js","assets/EnterpriseSettings-gbNle8-d.js","assets/SystemVerification-DzgCw_Vs.js","assets/ReferralsDashboard-DOFlTxTF.js","assets/EmailMonitor-DSeWNMg0.js","assets/ProspectTargets-BKn-muWC.js","assets/OutreachDashboard-D8D8aHio.js","assets/AuditLog-B2xLRYzM.js","assets/AiGovernance-Ce3GpJdR.js","assets/AiPerformance-BhOIkF_x.js","assets/WorkforceMarketplace-B6jDNEw4.js","assets/ConnectorHub-CSMei4X4.js","assets/categories-CmbRb4eK.js","assets/SchemaJsonLd-BeSR67e8.js","assets/ConnectorDetail-CjmkSnze.js","assets/EmployeeHub-DXNSMqdI.js","assets/EmployeeDetail-WqWsm6v-.js","assets/FirecrawlHealth-DGqRCjzU.js","assets/DeliverabilityDashboard-CJRWNJdk.js","assets/CampaignPreview-B1Oh3jRN.js"])))=>i.map(i=>d[i]); 2var ri=Object.defineProperty;var Pa=e=>{throw TypeError(e)};var ii=(e,t,a)=>t in e?ri(e,t,{enumerable:!0,configurable:!0,writable:!0,value:a}):e[t]=a;var qe=(e,t,a)=>ii(e,typeof t!="symbol"?t+"":t,a),On=(e,t,a)=>t.has(e)||Pa("Cannot "+a);var A=(e,t,a)=>(On(e,t,"read from private field"),a?a.call(e):t.get(e)),U=(e,t,a)=>t.has(e)?Pa("Cannot add the same private member more than once"):t instanceof WeakSet?t.add(e):t.set(e,a),F=(e,t,a,o)=>(On(e,t,"write to private field"),o?o.call(e,a):t.set(e,a),a),ae=(e,t,a)=>(On(e,t,"access private method"),a);var nn=(e,t,a,o)=>({set _(r){F(e,t,r,a)},get _(){return A(e,t,o)}});import{j as n,m as ct,A as si}from"./vendor-motion-VOv-vRjq.js";import{a as Co,r as i,R as C,b as li,g as la,u as Po,e as To,L as Ee,N as lt,B as ci,f as Ta,h as p}from"./vendor-react-CgwpgauT.js";import{c as di,a as ui,u as Eo,B as mi,P as Ft,V as Ro,b as pi,d as fi,e as bn,f as je,R as hi,g as gi,h as bi,i as yi,C as Oo,j as vi,k as Ii,T as Ai,S as xi,O as Mo,l as qo,m as wi,n as Do,D as Lo,o as Si,p as ki,q as Ci}from"./vendor-ui-kSdzIpai.js";import{c as Fo,R as Pi,a as Ti,P as Ei,b as Ri,d as Oi,e as Ea,L as Mi}from"./vendor-charts-BeJd8Fxm.js";import{u as Me,i as Qe,a as qi}from"./vendor-i18n-T2j15knT.js";import{c as Di}from"./vendor-supabase-DYXPcfum.js";(function(){const t=document.createElement("link").relList;if(t&&t.supports&&t.supports("modulepreload"))return;for(const r of document.querySelectorAll('link[rel="modulepreload"]'))o(r);new MutationObserver(r=>{for(const s of r)if(s.type==="childList")for(const d of s.addedNodes)d.tagName==="LINK"&&d.rel==="modulepreload"&&o(d)}).observe(document,{childList:!0,subtree:!0});function a(r){const s={};return r.integrity&&(s.integrity=r.integrity),r.referrerPolicy&&(s.referrerPolicy=r.referrerPolicy),r.crossOrigin==="use-credentials"?s.credentials="include":r.crossOrigin==="anonymous"?s.credentials="omit":s.credentials="same-origin",s}function o(r){if(r.ep)return;r.ep=!0;const s=a(r);fetch(r.href,s)}})();var No,Ra=Co;No=Ra.createRoot,Ra.hydrateRoot;const Li="modulepreload",Fi=function(e){return"/"+e},Oa={},f=function(t,a,o){let r=Promise.resolve();if(a&&a.length>0){document.getElementsByTagName("link");const d=document.querySelector("meta[property=csp-nonce]"),l=(d==null?void 0:d.nonce)||(d==null?void 0:d.getAttribute("nonce"));r=Promise.allSettled(a.map(m=>{if(m=Fi(m),m in Oa)return;Oa[m]=!0;const u=m.endsWith(".css"),v=u?'[rel="stylesheet"]':"";if(document.querySelector(`link[href="${m}"]${v}`))return;const h=document.createElement("link");
2if(h.rel=u?"stylesheet":Li,u||(h.as="script"),h.crossOrigin="",h.href=m,l&&h.setAttribute("nonce",l),document.head.appendChild(h),u)return new Promise((g,c)=>{h.addEventListener("load",g),h.addEventListener("error",()=>c(new Error(`Unable to preload CSS for ${m}`)))})}))}function s(d){const l=new Event("vite:preloadError",{cancelable:!0});if(l.payload=d,window.dispatchEvent(l),!l.defaultPrevented)throw d}return r.then(d=>{for(const l of d||[])l.status==="rejected"&&s(l.reason);return t().catch(s)})},Ni=1,_i=1e6;let Mn=0;function zi(){return Mn=(Mn+1)%Number.MAX_SAFE_INTEGER,Mn.toString()}const qn=new Map,Ma=e=>{if(qn.has(e))return;const t=setTimeout(()=>{qn.delete(e),Vt({type:"REMOVE_TOAST",toastId:e})},_i);qn.set(e,t)},ji=(e,t)=>{switch(t.type){case"ADD_TOAST":return{...e,toasts:[t.toast,...e.toasts].slice(0,Ni)};case"UPDATE_TOAST":return{...e,toasts:e.toasts.map(a=>a.id===t.toast.id?{...a,...t.toast}:a)};case"DISMISS_TOAST":{const{toastId:a}=t;return a?Ma(a):e.toasts.forEach(o=>{Ma(o.id)}),{...e,toasts:e.toasts.map(o=>o.id===a||a===void 0?{...o,open:!1}:o)}}case"REMOVE_TOAST":return t.toastId===void 0?{...e,toasts:[]}:{...e,toasts:e.toasts.filter(a=>a.id!==t.toastId)}}},mn=[];let pn={toasts:[]};
2function Vt(e){pn=ji(pn,e),mn.forEach(t=>{t(pn)})}function Wn({...e}){const t=zi(),a=r=>Vt({type:"UPDATE_TOAST",toast:{...r,id:t}}),o=()=>Vt({type:"DISMISS_TOAST",toastId:t});return Vt({type:"ADD_TOAST",toast:{...e,id:t,open:!0,onOpenChange:r=>{r||o()}}}),{id:t,dismiss:o,update:a}}function Gi(){const[e,t]=i.useState(pn);return i.useEffect(()=>(mn.push(t),()=>{const a=mn.indexOf(t);a>-1&&mn.splice(a,1)}),[e]),{...e,toast:Wn,dismiss:a=>Vt({type:"DISMISS_TOAST",toastId:a})}}var ca="ToastProvider",[da,Bi,Ui]=di("Toast"),[_o,Cg]=ui("Toast",[Ui]),[Qi,An]=_o(ca),zo=e=>{const{__scopeToast:t,label:a="Notification",duration:o=5e3,swipeDirection:r="right",swipeThreshold:s=50,children:d}=e,[l,m]=i.useState(null),[u,v]=i.useState(0),h=i.useRef(!1),g=i.useRef(!1);return a.trim()||console.error(`Invalid prop \`label\` supplied to \`${ca}\`. Expected non-empty \`string\`.`),n.jsx(da.Provider,{scope:t,children:n.jsx(Qi,{scope:t,label:a,duration:o,swipeDirection:r,swipeThreshold:s,toastCount:u,viewport:l,onViewportChange:m,onToastAdd:i.useCallback(()=>v(c=>c+1),[]),onToastRemove:i.useCallback(()=>v(c=>c-1),[]),isFocusedToastEscapeKeyDownRef:h,isClosePausedRef:g,children:d})})};zo.displayName=ca;var jo="ToastViewport",Vi=["F8"],Kn="toast.viewportPause",$n="toast.viewportResume",Go=i.forwardRef((e,t)=>{const{__scopeToast:a,hotkey:o=Vi,label:r="Notifications ({hotkey})",...s}=e,d=An(jo,a),l=Bi(a),m=i.useRef(null),u=i.useRef(null),v=i.useRef(null),h=i.useRef(null),g=Eo(t,h,d.onViewportChange),c=o.join("+").replace(/Key/g,"").replace(/Digit/g,""),w=d.toastCount>0;i.useEffect(()=>{const b=x=>{var k;o.length!==0&&o.every(P=>x[P]||x.code===P)&&((k=h.current)==null||k.focus())};return document.addEventListener("keydown",b),()=>document.removeEventListener("keydown",b)},[o]),i.useEffect(()=>{const b=m.current,x=h.current;if(w&&b&&x){const I=()=>{if(!d.isClosePausedRef.current){const E=new CustomEvent(Kn);x.dispatchEvent(E),d.isClosePausedRef.current=!0}},k=()=>{if(d.isClosePausedRef.current){const E=new CustomEvent($n);x.dispatchEvent(E),d.isClosePausedRef.current=!1}},P=E=>{!b.contains(E.relatedTarget)&&k()},T=()=>{b.contains(document.activeElement)||k()};return b.addEventListener("focusin",I),b.addEventListener("focusout",P),b.addEventListener("pointermove",I),b.addEventListener("pointerleave",T),window.addEventListener("blur",I),window.addEventListener("focus",k),()=>{b.removeEventListener("focusin",I),b.removeEventListener("focusout",P),b.removeEventListener("pointermove",I),b.removeEventListener("pointerleave",T),window.removeEventListener("blur",I),window.removeEventListener("focus",k)}}},[w,d.isClosePausedRef]);const y=i.useCallback(({tabbingDirection:b})=>{const I=l().map(k=>{const P=k.ref.current,T=[P,...os(P)];return b==="forwards"?T:T.reverse()});return(b==="forwards"?I.reverse():I).flat()},[l]);return i.useEffect(()=>{const b=h.current;if(b){const x=I=>{var T,E,j;const k=I.altKey||I.ctrlKey||I.metaKey;if(I.key==="Tab"&&!k){const D=document.activeElement,q=I.shiftKey;if(I.target===b&&q){(T=u.current)==null||T.focus();return}const S=y({tabbingDirection:q?"backwards":"forwards"}),M=S.findIndex(R=>R===D);Dn(S.slice(M+1))?I.preventDefault():q?(E=u.current)==null||E.focus():(j=v.current)==null||j.focus()}};return b.addEventListener("keydown",x),()=>b.removeEventListener("keydown",x)}},[l,y]),n.jsxs(mi,{ref:m,role:"region","aria-label":r.replace("{hotkey}",c),tabIndex:-1,style:{pointerEvents:w?void 0:"none"},children:[w&&n.jsx(Yn,{ref:u,onFocusFromOutsideViewport:()=>{const b=y({tabbingDirection:"forwards"});Dn(b)}}),n.jsx(da.Slot,{scope:a,children:n.jsx(Ft.ol,{tabIndex:-1,...s,ref:g})}),w&&n.jsx(Yn,{ref:v,onFocusFromOutsideViewport:()=>{const b=y({tabbingDirection:"backwards"});Dn(b)}})]})});Go.displayName=jo;var Bo="ToastFocusProxy",Yn=i.forwardRef((e,t)=>{const{__scopeToast:a,onFocusFromOutsideViewport:o,...r}=e,s=An(Bo,a);return n.jsx(Ro,{"aria-hidden":!0,tabIndex:0,...r,ref:t,style:{position:"fixed"},onFocus:d=>{var u;const l=d.relatedTarget;!((u=s.viewport)!=null&&u.contains(l))&&o()}})});Yn.displayName=Bo;var Jt="Toast",Hi="toast.swipeStart",Wi="toast.swipeMove",Ki="toast.swipeCancel",$i="toast.swipeEnd",Uo=i.forwardRef((e,t)=>
2{const{forceMount:a,open:o,defaultOpen:r,onOpenChange:s,...d}=e,[l,m]=pi({prop:o,defaultProp:r??!0,onChange:s,caller:Jt});return n.jsx(fi,{present:a||l,children:n.jsx(Ji,{open:l,...d,ref:t,onClose:()=>m(!1),onPause:bn(e.onPause),onResume:bn(e.onResume),onSwipeStart:je(e.onSwipeStart,u=>{u.currentTarget.setAttribute("data-swipe","start")}),onSwipeMove:je(e.onSwipeMove,u=>{const{x:v,y:h}=u.detail.delta;u.currentTarget.setAttribute("data-swipe","move"),u.currentTarget.style.setProperty("--radix-toast-swipe-move-x",`${v}px`),u.currentTarget.style.setProperty("--radix-toast-swipe-move-y",`${h}px`)}),onSwipeCancel:je(e.onSwipeCancel,u=>{u.currentTarget.setAttribute("data-swipe","cancel"),u.currentTarget.style.removeProperty("--radix-toast-swipe-move-x"),u.currentTarget.style.removeProperty("--radix-toast-swipe-move-y"),u.currentTarget.style.removeProperty("--radix-toast-swipe-end-x"),u.currentTarget.style.removeProperty("--radix-toast-swipe-end-y")}),onSwipeEnd:je(e.onSwipeEnd,u=>{const{x:v,y:h}=u.detail.delta;u.currentTarget.setAttribute("data-swipe","end"),u.currentTarget.style.removeProperty("--radix-toast-swipe-move-x"),u.currentTarget.style.removeProperty("--radix-toast-swipe-move-y"),u.currentTarget.style.setProperty("--radix-toast-swipe-end-x",`${v}px`),u.currentTarget.style.setProperty("--radix-toast-swipe-end-y",`${h}px`),m(!1)})})})});Uo.displayName=Jt;var[Yi,Zi]=_o(Jt,{onClose(){}}),Ji=i.forwardRef((e,t)=>{const{__scopeToast:a,type:o="foreground",duration:r,open:s,onClose:d,onEscapeKeyDown:l,onPause:m,onResume:u,onSwipeStart:v,onSwipeMove:h,onSwipeCancel:g,onSwipeEnd:c,...w}=e,y=An(Jt,a),[b,x]=i.useState(null),I=Eo(t,R=>x(R)),k=i.useRef(null),P=i.useRef(null),T=r||y.duration,E=i.useRef(0),j=i.useRef(T),D=i.useRef(0),{onToastAdd:q,onToastRemove:B}=y,G=bn(()=>{var Y;(b==null?void 0:b.contains(document.activeElement))&&((Y=y.viewport)==null||Y.focus()),d()}),S=i.useCallback(R=>{!R||R===1/0||(window.clearTimeout(D.current),E.current=new Date().getTime(),D.current=window.setTimeout(G,R))},[G]);i.useEffect(()=>{const R=y.viewport;if(R){const Y=()=>{S(j.current),u==null||u()},W=()=>{const ee=new Date().getTime()-E.current;j.current=j.current-ee,window.clearTimeout(D.current),m==null||m()};return R.addEventListener(Kn,W),R.addEventListener($n,Y),()=>{R.removeEventListener(Kn,W),R.removeEventListener($n,Y)}}},[y.viewport,T,m,u,S]),i.useEffect(()=>{s&&!y.isClosePausedRef.current&&S(T)},[s,T,y.isClosePausedRef,S]),i.useEffect(()=>(q(),()=>B()),[q,B]);const M=i.useMemo(()=>b?Yo(b):null,[b]);return y.viewport?n.jsxs(n.Fragment,{children:[M&&n.jsx(Xi,{__scopeToast:a,role:"status","aria-live":o==="foreground"?"assertive":"polite","aria-atomic":!0,children:M}),n.jsx(Yi,{scope:a,onClose:G,children:Co.createPortal(n.jsx(da.ItemSlot,{scope:a,children:n.jsx(hi,{asChild:!0,onEscapeKeyDown:je(l,()=>{y.isFocusedToastEscapeKeyDownRef.current||G(),y.isFocusedToastEscapeKeyDownRef.current=!1}),children:n.jsx(Ft.li,{role:"status","aria-live":"off","aria-atomic":!0,tabIndex:0,"data-state":s?"open":"closed","data-swipe-direction":y.swipeDirection,...w,ref:I,style:{userSelect:"none",touchAction:"none",...e.style},onKeyDown:je(e.onKeyDown,R=>{R.key==="Escape"&&(l==null||l(R.nativeEvent),R.nativeEvent.defaultPrevented||(y.isFocusedToastEscapeKeyDownRef.current=!0,G()))}),onPointerDown:je(e.onPointerDown,R=>{R.button===0&&(k.current={x:R.clientX,y:R.clientY})}),onPointerMove:je(e.onPointerMove,R=>{if(!k.current)return;const Y=R.clientX-k.current.x,W=R.clientY-k.current.y,ee=!!P.current,le=["left","right"].includes(y.swipeDirection),O=["left","up"].includes(y.swipeDirection)?Math.min:Math.max,ne=le?O(0,Y):0,ce=le?0:O(0,W),Q=R.pointerType==="touch"?10:2,V={x:ne,y:ce},K={originalEvent:R,delta:V};ee?(P.current=V,an(Wi,h,K,{discrete:!1})):qa(V,y.swipeDirection,Q)?(P.current=V,an(Hi,v,K,{discrete:!1}),R.target.setPointerCapture(R.pointerId)):(Math.abs(Y)>Q||Math.abs(W)>Q)&&(k.current=null)}),onPointerUp:je(e.onPointerUp,R=>{const Y=P.current,W=R.target;if(W.hasPointerCapture(R.pointerId)&&W.releasePointerCapture(R.pointerId),P.current=null,k.current=null,Y){const ee=R.currentTarget,le={originalEvent:R,delta:Y};qa(Y,y.swipeDirection,y.swipeThreshold)?an($i,c,le,{discrete:!0}):an(Ki,g,le,{discrete:!0}),ee.addEventListener("click",O=>O.preventDefault(),{once:!0})}})})})}),y.viewport)})]}):null}),Xi=e=>{const{__scopeToast:t,children:a,...o}=e,r=An(Jt,t),[s,d]=i.useState(!1),[l,m]=i.useState(!1);return ns(()=>d(!0)),i.useEffect(()=>{const u=window.setTimeout(()=>m(!0),1e3);return()=>window.clearTimeout(u)},[]),l?null:n.jsx(gi,{asChild:!0,children:n.jsx(Ro,{...o,children:s&&n.jsxs(n.Fragment,{children:[r.label," ",a]})})})},es="ToastTitle",Qo=i.forwardRef((e,t)=>
2{const{__scopeToast:a,...o}=e;return n.jsx(Ft.div,{...o,ref:t})});Qo.displayName=es;var ts="ToastDescription",Vo=i.forwardRef((e,t)=>{const{__scopeToast:a,...o}=e;return n.jsx(Ft.div,{...o,ref:t})});Vo.displayName=ts;var Ho="ToastAction",Wo=i.forwardRef((e,t)=>{const{altText:a,...o}=e;return a.trim()?n.jsx($o,{altText:a,asChild:!0,children:n.jsx(ua,{...o,ref:t})}):(console.error(`Invalid prop \`altText\` supplied to \`${Ho}\`. Expected non-empty \`string\`.`),null)});Wo.displayName=Ho;var Ko="ToastClose",ua=i.forwardRef((e,t)=>{const{__scopeToast:a,...o}=e,r=Zi(Ko,a);return n.jsx($o,{asChild:!0,children:n.jsx(Ft.button,{type:"button",...o,ref:t,onClick:je(e.onClick,r.onClose)})})});ua.displayName=Ko;var $o=i.forwardRef((e,t)=>{const{__scopeToast:a,altText:o,...r}=e;return n.jsx(Ft.div,{"data-radix-toast-announce-exclude":"","data-radix-toast-announce-alt":o||void 0,...r,ref:t})});function Yo(e){const t=[];return Array.from(e.childNodes).forEach(o=>{if(o.nodeType===o.TEXT_NODE&&o.textContent&&t.push(o.textContent),as(o)){const r=o.ariaHidden||o.hidden||o.style.display==="none",s=o.dataset.radixToastAnnounceExclude==="";if(!r)if(s){const d=o.dataset.radixToastAnnounceAlt;d&&t.push(d)}else t.push(...Yo(o))}}),t}function an(e,t,a,{discrete:o}){const r=a.originalEvent.currentTarget,s=new CustomEvent(e,{bubbles:!0,cancelable:!0,detail:a});t&&r.addEventListener(e,t,{once:!0}),o?yi(r,s):r.dispatchEvent(s)}var qa=(e,t,a=0)=>{const o=Math.abs(e.x),r=Math.abs(e.y),s=o>r;return t==="left"||t==="right"?s&&o>a:!s&&r>a};function ns(e=()=>{}){const t=bn(e);bi(()=>{let a=0,o=0;return a=window.requestAnimationFrame(()=>o=window.requestAnimationFrame(t)),()=>{window.cancelAnimationFrame(a),window.cancelAnimationFrame(o)}},[t])}function as(e){return e.nodeType===e.ELEMENT_NODE}function os(e){const t=[],a=document.createTreeWalker(e,NodeFilter.SHOW_ELEMENT,{acceptNode:o=>{const r=o.tagName==="INPUT"&&o.type==="hidden";return o.disabled||o.hidden||r?NodeFilter.FILTER_SKIP:o.tabIndex>=0?NodeFilter.FILTER_ACCEPT:NodeFilter.FILTER_SKIP}});for(;a.nextNode();)t.push(a.currentNode);return t}function Dn(e){const t=document.activeElement;return e.some(a=>a===t?!0:(a.focus(),document.activeElement!==t))}var rs=zo,Zo=Go,Jo=Uo,Xo=Qo,er=Vo,tr=Wo,nr=ua;const Da=e=>typeof e=="boolean"?`${e}`:e===0?"0":e,La=Fo,ar=(e,t)=>a=>{var o;if((t==null?void 0:t.variants)==null)return La(e,a==null?void 0:a.class,a==null?void 0:a.className);const{variants:r,defaultVariants:s}=t,d=Object.keys(r).map(u=>{const v=a==null?void 0:a[u],h=s==null?void 0:s[u];if(v===null)return null;const g=Da(v)||Da(h);return r[u][g]}),l=a&&Object.entries(a).reduce((u,v)=>{let[h,g]=v;return g===void 0||(u[h]=g),u},{}),m=t==null||(o=t.compoundVariants)===null||o===void 0?void 0:o.reduce((u,v)=>{let{class:h,className:g,...c}=v;return Object.entries(c).every(w=>{let[y,b]=w;return Array.isArray(b)?b.includes({...s,...l}[y]):{...s,...l}[y]===b})?[...u,h,g]:u},[]);return La(e,d,m,a==null?void 0:a.class,a==null?void 0:a.className)};/** 3 * @license lucide-react v0.462.0 - ISC 4 * 5 * This source code is licensed under the ISC license. 6 * See the LICENSE file in the root directory of this source tree. 7 */const is=e=>e.replace(/([a-z0-9])([A-Z])/g,"$1-$2").toLowerCase(),or=(...e)=>e.filter((t,a,o)=>!!t&&t.trim()!==""&&o.indexOf(t)===a).join(" ").trim();/** 8 * @license lucide-react v0.462.0 - ISC 9 * 10 * This source code is licensed under the ISC license. 11 * See the LICENSE file in the root directory of this source tree. 12 */var ss={xmlns:"http://www.w3.org/2000/svg",width:24,height:24,viewBox:"0 0 24 24",fill:"none",stroke:"currentColor",strokeWidth:2,strokeLinecap:"round",strokeLinejoin:"round"};/** 13 * @license lucide-react v0.462.0 - ISC 14 * 15 * This source code is licensed under the ISC license. 16 * See the LICENSE file in the root directory of this source tree. 17 */const ls=i.forwardRef(({color:e="currentColor",size:t=24,strokeWidth:a=2,absoluteStrokeWidth:o,className:r="",children:s,iconNode:d,...l},m)=>i.createElement("svg",{ref:m,...ss,width:t,height:t,stroke:e,strokeWidth:o?Number(a)*24/Number(t):a,className:or("lucide",r),...l},[...d.map(([u,v])=>i.createElement(u,v)),...Array.isArray(s)?s:[s]]));/** 18 * @license lucide-react v0.462.0 - ISC 19 * 20 * This source code is licensed under the ISC license. 21 * See the LICENSE file in the root directory of this source tree. 22 */const pe=(e,t)=>{const a=i.forwardRef(({className:o,...r},s)=>i.createElement(ls,{ref:s,iconNode:t,className:or(`lucide-${is(e)}`,o),...r}));return a.displayName=`${e}`,a};/** 23 * @license lucide-react v0.462.0 - ISC 24 * 25 * This source code is licensed under the ISC license. 26 * See the LICENSE file in the root directory of this source tree. 27 */const ma=pe("Activity",[["path",{d:"M22 12h-2.48a2 2 0 0 0-1.93 1.46l-2.35 8.36a.25.25 0 0 1-.48 0L9.24 2.18a.25.25 0 0 0-.48 0l
27-2.35 8.36A2 2 0 0 1 4.49 12H2",key:"169zse"}]]);/** 28 * @license lucide-react v0.462.0 - ISC 29 * 30 * This source code is licensed under the ISC license. 31 * See the LICENSE file in the root directory of this source tree. 32 */const cs=pe("ArrowRight",[["path",{d:"M5 12h14",key:"1ays0h"}],["path",{d:"m12 5 7 7-7 7",key:"xquz4c"}]]);/** 33 * @license lucide-react v0.462.0 - ISC 34 * 35 * This source code is licensed under the ISC license. 36 * See the LICENSE file in the root directory of this source tree. 37 */const ds=pe("Award",[["path",{d:"m15.477 12.89 1.515 8.526a.5.5 0 0 1-.81.47l-3.58-2.687a1 1 0 0 0-1.197 0l-3.586 2.686a.5.5 0 0 1-.81-.469l1.514-8.526",key:"1yiouv"}],["circle",{cx:"12",cy:"8",r:"6",key:"1vp47v"}]]);/** 38 * @license lucide-react v0.462.0 - ISC 39 * 40 * This source code is licensed under the ISC license. 41 * See the LICENSE file in the root directory of this source tree. 42 */const us=pe("Brain",[["path",{d:"M12 5a3 3 0 1 0-5.997.125 4 4 0 0 0-2.526 5.77 4 4 0 0 0 .556 6.588A4 4 0 1 0 12 18Z",key:"l5xja"}],["path",{d:"M12 5a3 3 0 1 1 5.997.125 4 4 0 0 1 2.526 5.77 4 4 0 0 1-.556 6.588A4 4 0 1 1 12 18Z",key:"ep3f8r"}],["path",{d:"M15 13a4.5 4.5 0 0 1-3-4 4.5 4.5 0 0 1-3 4",key:"1p4c4q"}],["path",{d:"M17.599 6.5a3 3 0 0 0 .399-1.375",key:"tmeiqw"}],["path",{d:"M6.003 5.125A3 3 0 0 0 6.401 6.5",key:"105sqy"}],["path",{d:"M3.477 10.896a4 4 0 0 1 .585-.396",key:"ql3yin"}],["path",{d:"M19.938 10.5a4 4 0 0 1 .585.396",key:"1qfode"}],["path",{d:"M6 18a4 4 0 0 1-1.967-.516",key:"2e4loj"}],["path",{d:"M19.967 17.484A4 4 0 0 1 18 18",key:"159ez6"}]]);/** 43 * @license lucide-react v0.462.0 - ISC 44 * 45 * This source code is licensed under the ISC license. 46 * See the LICENSE file in the root directory of this source tree. 47 */const ms=pe("Building",[["rect",{width:"16",height:"20",x:"4",y:"2",rx:"2",ry:"2",key:"76otgf"}],["path",{d:"M9 22v-4h6v4",key:"r93iot"}],["path",{d:"M8 6h.01",key:"1dz90k"}],["path",{d:"M16 6h.01",key:"1x0f13"}],["path",{d:"M12 6h.01",key:"1vi96p"}],["path",{d:"M12 10h.01",key:"1nrarc"}],["path",{d:"M12 14h.01",key:"1etili"}],["path",{d:"M16 10h.01",key:"1m94wz"}],["path",{d:"M16 14h.01",key:"1gbofw"}],["path",{d:"M8 10h.01",key:"19clt8"}],["path",{d:"M8 14h.01",key:"6423bh"}]]);/** 48 * @license lucide-react v0.462.0 - ISC 49 * 50 * This source code is licensed under the ISC license. 51 * See the LICENSE file in the root directory of this source tree. 52 */const ps=pe("ChevronDown",[["path",{d:"m6 9 6 6 6-6",key:"qrunsl"}]]);/** 53 * @license lucide-react v0.462.0 - ISC 54 * 55 * This source code is licensed under the ISC license. 56 * See the LICENSE file in the root directory of this source tree. 57 */const fs=pe("CircleCheckBig",[["path",{d:"M21.801 10A10 10 0 1 1 17 3.335",key:"yps3ct"}],["path",{d:"m9 11 3 3L22 4",key:"1pflzl"}]]);/** 58 * @license lucide-react v0.462.0 - ISC 59 * 60 * This source code is licensed under the ISC license. 61 * See the LICENSE file in the root directory of this source tree. 62 */const hs=pe("ExternalLink",[["path",{d:"M15 3h6v6",key:"1q9fwt"}],["path",{d:"M10 14 21 3",key:"gplh6r"}],["path",{d:"M18 13v6a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h6",key:"a6xqqp"}]]);/** 63 * @license lucide-react v0.462.0 - ISC 64 * 65 * This source code is licensed under the ISC license. 66 * See the LICENSE file in the root directory of this source tree. 67 */const rr=pe("Lock",[["rect",{width:"18",height:"11",x:"3",y:"11",rx:"2",ry:"2",key:"1w4ew1"}],["path",{d:"M7 11V7a5 5 0 0 1 10 0v4",key:"fwvmzm"}]]);/** 68 * @license lucide-react v0.462.0 - ISC 69 * 70 * This source code is licensed under the ISC license. 71 * See the LICENSE file in the root directory of this source tree. 72 */const gs=pe("Mail",[["rect",{width:"20",height:"16",x:"2",y:"4",rx:"2",key:"18n3k1"}],["path",{d:"m22 7-8.97 5.7a1.94 1.94 0 0 1-2.06 0L2 7",key:"1ocrg3"}]]);/** 73 * @license lucide-react v0.462.0 - ISC 74 * 75 * This source code is licensed under the ISC license. 76 * See the LICENSE file in the root directory of this source tree. 77 */const bs=pe("Menu",[["line",{x1:"4",x2:"20",y1:"12",y2:"12",key:"1e0a9i"}],["line",{x1:"4",x2:"20",y1:"6",y2:"6",key:"1owob3"}],["line",{x1:"4",x2:"20",y1:"18",y2:"18",key:"yk5zj1"}]]);/** 78 * @license lucide-react v0.462.0 - ISC 79 * 80 * This source code is licensed under the ISC license. 81 * See the LICENSE file in the root directory of this source tree. 82 */const ir=pe("Shield",[["path",{d:"M20 13c0 5-3.5 7.5-7.66 8.95a1 1 0 0 1-.67-.01C7.5 20.5 4 18 4 13V6a1 1 0 0 1 1-1c2 0 4.5-1.2 6.24-2.72a1.17 1.17 0 0 1 1.52 0C14.51 3.81 17 5 19 5a1 1 0 0 1 1 1z",ke
vendor: 8,395 bytes, lines 82-97
82y:"oel41y"}]]);/** 83 * @license lucide-react v0.462.0 - ISC 84 * 85 * This source code is licensed under the ISC license. 86 * See the LICENSE file in the root directory of this source tree. 87 */const ys=pe("User",[["path",{d:"M19 21v-2a4 4 0 0 0-4-4H9a4 4 0 0 0-4 4v2",key:"975kel"}],["circle",{cx:"12",cy:"7",r:"4",key:"17ys0d"}]]);/** 88 * @license lucide-react v0.462.0 - ISC 89 * 90 * This source code is licensed under the ISC license. 91 * See the LICENSE file in the root directory of this source tree. 92 */const xn=pe("X",[["path",{d:"M18 6 6 18",key:"1bl5f8"}],["path",{d:"m6 6 12 12",key:"d8bk6v"}]]);/** 93 * @license lucide-react v0.462.0 - ISC 94 * 95 * This source code is licensed under the ISC license. 96 * See the LICENSE file in the root directory of this source tree. 97 */const vs=pe("Zap",[["path",{d:"M4 14a1 1 0 0 1-.78-1.63l9.9-10.2a.5.5 0 0 1 .86.46l-1.92 6.02A1 1 0 0 0 13 10h7a1 1 0 0 1 .78 1.63l-9.9 10.2a.5.5 0 0 1-.86-.46l1.92-6.02A1 1 0 0 0 11 14z",key:"1xq2db"}]]),pa="-",Is=e=>{const t=xs(e),{conflictingClassGroups:a,conflictingClassGroupModifiers:o}=e;return{getClassGroupId:d=>{const l=d.split(pa);return l[0]===""&&l.length!==1&&l.shift(),sr(l,t)||As(d)},getConflictingClassGroupIds:(d,l)=>{const m=a[d]||[];return l&&o[d]?[...m,...o[d]]:m}}},sr=(e,t)=>{var d;if(e.length===0)return t.classGroupId;const a=e[0],o=t.nextPart.get(a),r=o?sr(e.slice(1),o):void 0;if(r)return r;if(t.validators.length===0)return;const s=e.join(pa);return(d=t.validators.find(({validator:l})=>l(s)))==null?void 0:d.classGroupId},Fa=/^\[(.+)\]$/,As=e=>{if(Fa.test(e)){const t=Fa.exec(e)[1],a=t==null?void 0:t.substring(0,t.indexOf(":"));if(a)return"arbitrary.."+a}},xs=e=>{const{theme:t,prefix:a}=e,o={nextPart:new Map,validators:[]};return Ss(Object.entries(e.classGroups),a).forEach(([s,d])=>{Zn(d,o,s,t)}),o},Zn=(e,t,a,o)=>{e.forEach(r=>{if(typeof r=="string"){const s=r===""?t:Na(t,r);s.classGroupId=a;return}if(typeof r=="function"){if(ws(r)){Zn(r(o),t,a,o);return}t.validators.push({validator:r,classGroupId:a});return}Object.entries(r).forEach(([s,d])=>{Zn(d,Na(t,s),a,o)})})},Na=(e,t)=>{let a=e;return t.split(pa).forEach(o=>{a.nextPart.has(o)||a.nextPart.set(o,{nextPart:new Map,validators:[]}),a=a.nextPart.get(o)}),a},ws=e=>e.isThemeGetter,Ss=(e,t)=>t?e.map(([a,o])=>{const r=o.map(s=>typeof s=="string"?t+s:typeof s=="object"?Object.fromEntries(Object.entries(s).map(([d,l])=>[t+d,l])):s);return[a,r]}):e,ks=e=>{if(e<1)return{get:()=>{},set:()=>{}};let t=0,a=new Map,o=new Map;const r=(s,d)=>{a.set(s,d),t++,t>e&&(t=0,o=a,a=new Map)};return{get(s){let d=a.get(s);if(d!==void 0)return d;if((d=o.get(s))!==void 0)return r(s,d),d},set(s,d){a.has(s)?a.set(s,d):r(s,d)}}},lr="!",Cs=e=>{const{separator:t,experimentalParseClassName:a}=e,o=t.length===1,r=t[0],s=t.length,d=l=>{const m=[];let u=0,v=0,h;for(let b=0;b<l.length;b++){let x=l[b];if(u===0){if(x===r&&(o||l.slice(b,b+s)===t)){m.push(l.slice(v,b)),v=b+s;continue}if(x==="/"){h=b;continue}}x==="["?u++:x==="]"&&u--}const g=m.length===0?l:l.substring(v),c=g.startsWith(lr),w=c?g.substring(1):g,y=h&&h>v?h-v:void 0;return{modifiers:m,hasImportantModifier:c,baseClassName:w,maybePostfixModifierPosition:y}};return a?l=>a({className:l,parseClassName:d}):d},Ps=e=>{if(e.length<=1)return e;const t=[];let a=[];return e.forEach(o=>{o[0]==="["?(t.push(...a.sort(),o),a=[]):a.push(o)}),t.push(...a.sort()),t},Ts=e=>({cache:ks(e.cacheSize),parseClassName:Cs(e),...Is(e)}),Es=/\s+/,Rs=(e,t)=>{const{parseClassName:a,getClassGroupId:o,getConflictingClassGroupIds:r}=t,s=[],d=e.trim().split(Es);let l="";for(let m=d.length-1;m>=0;m-=1){const u=d[m],{modifiers:v,hasImportantModifier:h,baseClassName:g,maybePostfixModifierPosition:c}=a(u);let w=!!c,y=o(w?g.substring(0,c):g);if(!y){if(!w){l=u+(l.length>0?" "+l:l);continue}if(y=o(g),!y){l=u+(l.length>0?" "+l:l);continue}w=!1}const b=Ps(v).join(":"),x=h?b+lr:b,I=x+y;if(s.includes(I))continue;s.push(I);const k=r(y,w);for(let P=0;P<k.length;++P){const T=k[P];s.push(x+T)}l=u+(l.length>0?" "+l:l)}return l};function Os(){let e=0,t,a,o="";for(;e<arguments.length;)(t=arguments[e++])&&(a=cr(t))&&(o&&(o+=" "),o+=a);return o}const cr=e=>{if(typeof e=="string")return e;let t,a="";for(let o=0;o<e.length;o++)e[o]&&(t=cr(e[o]))&&(a&&(a+=" "),a+=t);return a};function Ms(e,...t){let a,o,r,s=d;function d(m){const u=t.reduce((v,h)=>h(v),e());return a=Ts(u),o=a.cache.get,r=a.cache.set,s=l,l(m)}function l(m){const u=o(m);if(u)return u;const v=Rs(m,a);return r(m,v),v}return function(){return s(Os.apply(null,arguments))}}const $=e=>{const t=a=>a[e]||[];return t.isThemeGetter=!0,t},dr=/^\[(?:([a-z-]+):)?(.+)\]$/i,qs=/^\d+\/\d+$/,Ds=new Set(["px","full","screen"]),Ls=/^(\d+(\.\d+)?)?(xs|sm|md|lg|xl)$/,Fs=/\d+(%|px|r?em|[sdl]?v([hwib]|min|max)|pt|pc|in|cm|mm|cap|ch|ex|r?lh|cq(w|h|i|b|min|max))|\b(calc|min|max|clamp)\(.+\)|^0$/,Ns=/^(rgba?|hsla?|hwb|(ok)?(lab|lch))\(.+\)$/,_s=/^(inset_)?-?((\d+)?\.?(\d+)[a-z]+|0)_-?((\d+)?\.?(\d+)[a-z]+|0)/,zs=/^(url|image|image-set|cross-fade|element|(repeating-)?(linear|radial|conic)-gradient)\(.+\)$/,Ge=e=>St(e)||Ds.has(e)||qs.test(e),$e=e=>Nt(e,"length",Ws),St=e=>!!e&&!Number.isNaN(Number(e)),Ln=e=>Nt(e,"number",St),Gt=e=>!!e&&Number.isInteger(Number(e)),js=e=>e.endsWith("%")&&St(e.slice(0,-1)),z=e=>dr.test(e),Ye=e=>Ls.test(e),Gs=new Set(["length","size","percentage"]),Bs=e=>Nt(e,Gs,ur),Us=e=>Nt(e,"position",ur),Qs=new Set(["image","url"]),Vs=e=>Nt(e,Qs,$s),Hs=e=>Nt(e,"",Ks),Bt=()=>!0,Nt=(e,t,a)=>{const o=dr.exec(e);return o?o[1]?typeof t=="string"?o[1]===t:t.has(o[1]):a(o[2]):!1},Ws=e=>Fs.test(e)&&!Ns.test(e),ur=()=>!1,Ks=e=>_s.test(e),$s=e=>zs.test(e),Ys=()=>{const e=$("colors"),t=$("spacing"),a=$("blur"),o=$("brightness"),r=$("borderColor"),s=$("borderRadius"),d=$("borderSpacing"),l=$("borderWidth"),m=$("contrast"),u=$("grayscale"),v=$("hueRotate"),h=$("invert"),g=$("gap"),c=$("gradientColorStops"),w=$("gradientColorStopPositions"),y=$("inset"),b=$("margin"),x=$("opacity"),I=$("padding"),k=$("saturate"),P=$("scale"),T=$("sepia"),E=$("skew"),j=$("space"),D=$("translate"),q=()=>["auto","contain","none"],B=()=>["auto","hidden","clip","visible","scroll"],G=()=>["auto",z,t],S=()=>[z,t],M=()=>["",Ge,$e],R=()=>["auto",St,z],Y=()=>["bottom","center","left","left-bottom","left-top","right","right-bottom","right-top","top"],W=()=>["solid","dashed","dotted","double","none"],ee=()=>["normal","multiply","screen","overlay","darken","lighten","color-dodge","color-burn","hard-light","soft-light","difference","exclusion","hue","saturation","color","luminosity"],le=()=>["start","end","center","between","around","evenly","stretch"],O=()=>["","0",z],ne=()=>["auto","avoid","all","avoid-page","page","left","right","column"],ce=()=>[St,z];return{cacheSize:500,separator:":",theme:{colors:[Bt],spacing:[Ge,$e],blur:["none","",Ye,z],brightness:ce(),borderColor:[e],borderRadius:["none","","full",Ye,z],borderSpacing:S(),borderWidth:M(),contrast:ce(),grayscale:O(),hueRotate:ce(),invert:O(),gap:S(),gradientColorStops:[e],gradientColorStopPositions:[js,$e],inset:G(),margin:G(),opacity:ce(),padding:S(),saturate:ce(),scale:ce(),sepia:O(),skew:ce(),space:S(),translate:S()},classGroups:{aspect:[{aspect:["auto","square","video",z]}],container:["container"],columns:[{columns:[Ye]}],"break-after":[{"break-after":ne()}],"break-before":[{"break-before":ne()}],"break-inside":[{"break-inside":["auto","avoid","avoid-page","avoid-column"]}],"box-decoration":[{"box-decoration":["slice","clone"]}],box:[{box:["border","content"]}],display:["block","inline-block","inline","flex","inline-flex","table","inline-table","table-caption","table-cell","table-column","table-column-group","table-footer-group","table-header-group","table-row-group","table-row","flow-root","grid","inline-grid","contents","list-item","hidden"],float:[{float:["right","left","none","start","end"]}],clear:[{clear:["left","right","both","none","start","end"]}],isolation:["isolate","isolation-auto"],"object-fit":[{object:["contain","cover","fill","none","scale-down"]}],"object-position":[{object:[...Y(),z]}],overflow:[{overflow:B()}],"overflow-x":[{"overflow-x":B()}],"overflow-y":[{"overflow-y":B()}],overscroll:[{overscroll:q()}],"overscroll-x":[{"overscroll-x":q()}],"overscroll-y":[{"overscroll-y":q()}],position:["static","fixed","absolute","relative","sticky"],inset:[{inset:[y]}],"inset-x":[{"inset-x":[y]}],"inset-y":[{"inset-y":[y]}],start:[{start:[y]}],end:[{end:[y]}],top:[{top:[y]}
vendor: 8,668 bytes, line 97
97],right:[{right:[y]}],bottom:[{bottom:[y]}],left:[{left:[y]}],visibility:["visible","invisible","collapse"],z:[{z:["auto",Gt,z]}],basis:[{basis:G()}],"flex-direction":[{flex:["row","row-reverse","col","col-reverse"]}],"flex-wrap":[{flex:["wrap","wrap-reverse","nowrap"]}],flex:[{flex:["1","auto","initial","none",z]}],grow:[{grow:O()}],shrink:[{shrink:O()}],order:[{order:["first","last","none",Gt,z]}],"grid-cols":[{"grid-cols":[Bt]}],"col-start-end":[{col:["auto",{span:["full",Gt,z]},z]}],"col-start":[{"col-start":R()}],"col-end":[{"col-end":R()}],"grid-rows":[{"grid-rows":[Bt]}],"row-start-end":[{row:["auto",{span:[Gt,z]},z]}],"row-start":[{"row-start":R()}],"row-end":[{"row-end":R()}],"grid-flow":[{"grid-flow":["row","col","dense","row-dense","col-dense"]}],"auto-cols":[{"auto-cols":["auto","min","max","fr",z]}],"auto-rows":[{"auto-rows":["auto","min","max","fr",z]}],gap:[{gap:[g]}],"gap-x":[{"gap-x":[g]}],"gap-y":[{"gap-y":[g]}],"justify-content":[{justify:["normal",...le()]}],"justify-items":[{"justify-items":["start","end","center","stretch"]}],"justify-self":[{"justify-self":["auto","start","end","center","stretch"]}],"align-content":[{content:["normal",...le(),"baseline"]}],"align-items":[{items:["start","end","center","baseline","stretch"]}],"align-self":[{self:["auto","start","end","center","stretch","baseline"]}],"place-content":[{"place-content":[...le(),"baseline"]}],"place-items":[{"place-items":["start","end","center","baseline","stretch"]}],"place-self":[{"place-self":["auto","start","end","center","stretch"]}],p:[{p:[I]}],px:[{px:[I]}],py:[{py:[I]}],ps:[{ps:[I]}],pe:[{pe:[I]}],pt:[{pt:[I]}],pr:[{pr:[I]}],pb:[{pb:[I]}],pl:[{pl:[I]}],m:[{m:[b]}],mx:[{mx:[b]}],my:[{my:[b]}],ms:[{ms:[b]}],me:[{me:[b]}],mt:[{mt:[b]}],mr:[{mr:[b]}],mb:[{mb:[b]}],ml:[{ml:[b]}],"space-x":[{"space-x":[j]}],"space-x-reverse":["space-x-reverse"],"space-y":[{"space-y":[j]}],"space-y-reverse":["space-y-reverse"],w:[{w:["auto","min","max","fit","svw","lvw","dvw",z,t]}],"min-w":[{"min-w":[z,t,"min","max","fit"]}],"max-w":[{"max-w":[z,t,"none","full","min","max","fit","prose",{screen:[Ye]},Ye]}],h:[{h:[z,t,"auto","min","max","fit","svh","lvh","dvh"]}],"min-h":[{"min-h":[z,t,"min","max","fit","svh","lvh","dvh"]}],"max-h":[{"max-h":[z,t,"min","max","fit","svh","lvh","dvh"]}],size:[{size:[z,t,"auto","min","max","fit"]}],"font-size":[{text:["base",Ye,$e]}],"font-smoothing":["antialiased","subpixel-antialiased"],"font-style":["italic","not-italic"],"font-weight":[{font:["thin","extralight","light","normal","medium","semibold","bold","extrabold","black",Ln]}],"font-family":[{font:[Bt]}],"fvn-normal":["normal-nums"],"fvn-ordinal":["ordinal"],"fvn-slashed-zero":["slashed-zero"],"fvn-figure":["lining-nums","oldstyle-nums"],"fvn-spacing":["proportional-nums","tabular-nums"],"fvn-fraction":["diagonal-fractions","stacked-fractions"],tracking:[{tracking:["tighter","tight","normal","wide","wider","widest",z]}],"line-clamp":[{"line-clamp":["none",St,Ln]}],leading:[{leading:["none","tight","snug","normal","relaxed","loose",Ge,z]}],"list-image":[{"list-image":["none",z]}],"list-style-type":[{list:["none","disc","decimal",z]}],"list-style-position":[{list:["inside","outside"]}],"placeholder-color":[{placeholder:[e]}],"placeholder-opacity":[{"placeholder-opacity":[x]}],"text-alignment":[{text:["left","center","right","justify","start","end"]}],"text-color":[{text:[e]}],"text-opacity":[{"text-opacity":[x]}],"text-decoration":["underline","overline","line-through","no-underline"],"text-decoration-style":[{decoration:[...W(),"wavy"]}],"text-decoration-thickness":[{decoration:["auto","from-font",Ge,$e]}],"underline-offset":[{"underline-offset":["auto",Ge,z]}],"text-decoration-color":[{decoration:[e]}],"text-transform":["uppercase","lowercase","capitalize","normal-case"],"text-overflow":["truncate","text-ellipsis","text-clip"],"text-wrap":[{text:["wrap","nowrap","balance","pretty"]}],indent:[{indent:S()}],"vertical-align":[{align:["baseline","top","middle","bottom","text-top","text-bottom","sub","super",z]}],whitespace:[{whitespace:["normal","nowrap","pre","pre-line","pre-wrap","break-spaces"]}],break:[{break:["normal","words","all","keep"]}],hyphens:[{hyphens:["none","manual","auto"]}],content:[{content:["none",z]}],"bg-attachment":[{bg:["fixed","local","scroll"]}],"bg-clip":[{"bg-clip":["border","padding","content","text"]}],"bg-opacity":[{"bg-opacity":[x]}],"bg-origin":[{"bg-origin":["border","padding","content"]}],"bg-position":[{bg:[...Y(),Us]}],"bg-repeat":[{bg:["no-repeat",{repeat:["","x","y","round","space"]}]}],"bg-size":[{bg:["auto","cover","contain",Bs]}],"bg-image":[{bg:["none",{"gradient-to":["t","tr","r","br","b","bl","l","tl"]},Vs]}],"bg-color":[{bg:[e]}],"gradient-from-pos":[{from:[w]}],"gradient-via-pos":[{via:[w]}],"gradient-to-pos":[{to:[w]}],"gradient-from":[{from:[c]}],"gradient-via":[{via:[c]}],"gradient-to":[{to:[c]}],rounded:[{rounded:[s]}],"rounded-s":[{"rounded-s":[s]}],"rounded-e":[{"rounded-e":[s]}],"rounded-t":[{"rounded-t":[s]}],"rounded-r":[{"rounded-r":[s]}],"rounded-b":[{"rounded-b":[s]}],"rounded-l":[{"rounded-l":[s]}],"rounded-ss":[{"rounded-ss":[s]}],"rounded-se":[{"rounded-se":[s]}],"rounded-ee":[{"rounded-ee":[s]}],"rounded-es":[{"rounded-es":[s]}],"rounded-tl":[{"rounded-tl":[s]}],"rounded-tr":[{"rounded-tr":[s]}],"rounded-br":[{"rounded-br":[s]}],"rounded-bl":[{"rounded-bl":[s]}],"border-w":[{border:[l]}],"border-w-x":[{"border-x":[l]}],"border-w-y":[{"border-y":[l]}],"border-w-s":[{"border-s":[l]}],"border-w-e":[{"border-e":[l]}],"border-w-t":[{"border-t":[l]}],"border-w-r":[{"border-r":[l]}],"border-w-b":[{"border-b":[l]}],"border-w-l":[{"border-l":[l]}],"border-opacity":[{"border-opacity":[x]}],"border-style":[{border:[...W(),"hidden"]}],"divide-x":[{"divide-x":[l]}],"divide-x-reverse":["divide-x-reverse"],"divide-y":[{"divide-y":[l]}],"divide-y-reverse":["divide-y-reverse"],"divide-opacity":[{"divide-opacity":[x]}],"divide-style":[{divide:W()}],"border-color":[{border:[r]}],"border-color-x":[{"border-x":[r]}],"border-color-y":[{"border-y":[r]}],"border-color-s":[{"border-s":[r]}],"border-color-e":[{"border-e":[r]}],"border-color-t":[{"border-t":[r]}],"border-color-r":[{"border-r":[r]}],"border-color-b":[{"border-b":[r]}],"border-color-l":[{"border-l":[r]}],"divide-color":[{divide:[r]}],"outline-style":[{outline:["",...W()]}],"outline-offset":[{"outline-offset":[Ge,z]}],"outline-w":[{outline:[Ge,$e]}],"outline-color":[{outline:[e]}],"ring-w":[{ring:M()}],"ring-w-inset":["ring-inset"],"ring-color":[{ring:[e]}],"ring-opacity":[{"ring-opacity":[x]}],"ring-offset-w":[{"ring-offset":[Ge,$e]}],"ring-offset-color":[{"ring-offset":[e]}],shadow:[{shadow:["","inner","none",Ye,Hs]}],"shadow-color":[{shadow:[Bt]}],opacity:[{opacity:[x]}],"mix-blend":[{"mix-blend":[...ee(),"plus-lighter","plus-darker"]}],"bg-blend":[{"bg-blend":ee()}],filter:[{filter:["","none"]}],blur:[{blur:[a]}],brightness:[{brightness:[o]}],contrast:[{contrast:[m]}],"drop-shadow":[{"drop-shadow":["","none",Ye,z]}],grayscale:[{grayscale:[u]}],"hue-rotate":[{"hue-rotate":[v]}],invert:[{invert:[h]}],saturate:[{saturate:[k]}],sepia:[{sepia:[T]}],"backdrop-filter":[{"backdrop-filter":["","none"]}],"backdrop-blur":[{"backdrop-blur":[a]}],"backdrop-brightness":[{"backdrop-brightness":[o]}],"backdrop-contrast":[{"backdrop-contrast":[m]}],"backdrop-grayscale":[{"backdrop-grayscale":[u]}],"backdrop-hue-rotate":[{"backdrop-hue-rotate":[v]}],"backdrop-invert":[{"backdrop-invert":[h]}],"backdrop-opacity":[{"backdrop-opacity":[x]}],"backdrop-saturate":[{"backdrop-saturate":[k]}],"backdrop-sepia":[{"backdrop-sepia":[T]}],"border-collapse":[{border:["collapse","separate"]}],"border-spacing":[{"border-spacing":[d]}],"border-spacing-x":[{"border-spacing-x":[d]}],"border-spacing-y":[{"border-spacing-y":[d]}],"table-layout":[{table:["auto","fixed"]}],caption:[{caption:["top","bottom"]}],transition:[{transition:["none","all","","colors","opacity","shadow","transform",z]}],duration:[{duration:ce()}],ease:[{ease:["linear","in","out","in-out",z]}],delay:[{delay:ce()}],animate:[{animate:["none","spin","ping","pulse","bounce",z]}],transform:[{transform:["","gpu","none"]}],scale:[{scale:[P]}],"scale-x":[{"scale-x":[P]}],"scale-y":[{"scale-y":[P]}],rotate:[{rotate:[Gt,z]}],"translate-x":[{"translate-x":[D]}],"translate-y":[{"translate-y":[D]}],"skew-x":[{"skew-x":[E]}],"skew-y":[{"skew-y":[E]}],"transform-origin":[{origin:["center","top","top-right","right","bottom-right","bottom","bottom-left","left","top-left",z]}],accent:[{accent:["auto",e]}],appearance:[{appearance:["none","auto"]}],cursor:[{cursor:["auto","default","pointer","wait","text","move","help","not-allowed","
97none","context-menu","progress","cell","crosshair","vertical-text","alias","copy","no-drop","grab","grabbing","all-scroll","col-resize","row-resize","n-resize","e-resize","s-resize","w-resize","ne-resize","nw-resize","se-resize","sw-resize","ew-resize","ns-resize","nesw-resize","nwse-resize","zoom-in","zoom-out",z]}],"caret-color":[{caret:[e]}],"pointer-events":[{"pointer-events":["none","auto"]}],resize:[{resize:["none","y","x",""]}],"scroll-behavior":[{scroll:["auto","smooth"]}],"scroll-m":[{"scroll-m":S()}],"scroll-mx":[{"scroll-mx":S()}],"scroll-my":[{"scroll-my":S()}],"scroll-ms":[{"scroll-ms":S()}],"scroll-me":[{"scroll-me":S()}],"scroll-mt":[{"scroll-mt":S()}],"scroll-mr":[{"scroll-mr":S()}],"scroll-mb":[{"scroll-mb":S()}],"scroll-ml":[{"scroll-ml":S()}],"scroll-p":[{"scroll-p":S()}],"scroll-px":[{"scroll-px":S()}],"scroll-py":[{"scroll-py":S()}],"scroll-ps":[{"scroll-ps":S()}],"scroll-pe":[{"scroll-pe":S()}],"scroll-pt":[{"scroll-pt":S()}],"scroll-pr":[{"scroll-pr":S()}],"scroll-pb":[{"scroll-pb":S()}],"scroll-pl":[{"scroll-pl":S()}],"snap-align":[{snap:["start","end","center","align-none"]}],"snap-stop":[{snap:["normal","always"]}],"snap-type":[{snap:["none","x","y","both"]}],"snap-strictness":[{snap:["mandatory","proximity"]}],touch:[{touch:["auto","none","manipulation"]}],"touch-x":[{"touch-pan":["x","left","right"]}],"touch-y":[{"touch-pan":["y","up","down"]}],"touch-pz":["touch-pinch-zoom"],select:[{select:["none","text","all","auto"]}],"will-change":[{"will-change":["auto","scroll","contents","transform",z]}],fill:[{fill:[e,"none"]}],"stroke-w":[{stroke:[Ge,$e,Ln]}],stroke:[{stroke:[e,"none"]}],sr:["sr-only","not-sr-only"],"forced-color-adjust":[{"forced-color-adjust":["auto","none"]}]},conflictingClassGroups:{overflow:["overflow-x","overflow-y"],overscroll:["overscroll-x","overscroll-y"],inset:["inset-x","inset-y","start","end","top","right","bottom","left"],"inset-x":["right","left"],"inset-y":["top","bottom"],flex:["basis","grow","shrink"],gap:["gap-x","gap-y"],p:["px","py","ps","pe","pt","pr","pb","pl"],px:["pr","pl"],py:["pt","pb"],m:["mx","my","ms","me","mt","mr","mb","ml"],mx:["mr","ml"],my:["mt","mb"],size:["w","h"],"font-size":["leading"],"fvn-normal":["fvn-ordinal","fvn-slashed-zero","fvn-figure","fvn-spacing","fvn-fraction"],"fvn-ordinal":["fvn-normal"],"fvn-slashed-zero":["fvn-normal"],"fvn-figure":["fvn-normal"],"fvn-spacing":["fvn-normal"],"fvn-fraction":["fvn-normal"],"line-clamp":["display","overflow"],rounded:["rounded-s","rounded-e","rounded-t","rounded-r","rounded-b","rounded-l","rounded-ss","rounded-se","rounded-ee","rounded-es","rounded-tl","rounded-tr","rounded-br","rounded-bl"],"rounded-s":["rounded-ss","rounded-es"],"rounded-e":["rounded-se","rounded-ee"],"rounded-t":["rounded-tl","rounded-tr"],"rounded-r":["rounded-tr","rounded-br"],"rounded-b":["rounded-br","rounded-bl"],"rounded-l":["rounded-tl","rounded-bl"],"border-spacing":["border-spacing-x","border-spacing-y"],"border-w":["border-w-s","border-w-e","border-w-t","border-w-r","border-w-b","border-w-l"],"border-w-x":["border-w-r","border-w-l"],"border-w-y":["border-w-t","border-w-b"],"border-color":["border-color-s","border-color-e","border-color-t","border-color-r","border-color-b","border-color-l"],"border-color-x":["border-color-r","border-color-l"],"border-color-y":["border-color-t","border-color-b"],"scroll-m":["scroll-mx","scroll-my","scroll-ms","scroll-me","scroll-mt","scroll-mr","scroll-mb","scroll-ml"],"scroll-mx":["scroll-mr","scroll-ml"],"scroll-my":["scroll-mt","scroll-mb"],"scroll-p":["scroll-px","scroll-py","scroll-ps","scroll-pe","scroll-pt","scroll-pr","scroll-pb","scroll-pl"],"scroll-px":["scroll-pr","scroll-pl"],"scroll-py":["scroll-pt","scroll-pb"],touch:["touch-x","touch-y","touch-pz"],"touch-x":["touch"],"touch-y":["touch"],"touch-pz":["touch"]},conflictingClassGroupModifiers:{"font-size":["leading"]}}},Zs=Ms(Ys);function se(...e){return Zs(Fo(e))}const Js=rs,mr=i.forwardRef(({className:e,...t},a)=>n.jsx(Zo,{ref:a,className:se("fixed top-0 z-[100] flex max-h-screen w-full flex-col-reverse p-4 sm:bottom-0 sm:right-0 sm:top-auto sm:flex-col md:max-w-[420px]",e),...t}));mr.displayName=Zo.displayName;const Xs=ar("group pointer-events-auto relative flex w-full items-center justify-between space-x-4 overflow-hidden rounded-md border p-6 pr-8 shadow-lg transition-all data-[swipe=cancel]:translate-x-0 data-[swipe=end]:translate-x-[var(--radix-toast-swipe-end-x)] data-[swipe=move]:translate-x-[var(--radix-toast-swipe-move-x)] data-[swipe=move]:transition-none data-[state=open]:animate-in data-[state=closed]:animate-out data-[swipe=end]:animate-out data-[state=closed]:fade-out-80 data-[state=closed]:slide-out-to-right-full data-[state=open]:slide-in-from-top-full data-[state=open]:sm:slide-in-from-bottom-full",{variants:{variant:{default:"border bg-background text-foreground",destructive:"destructive group border-destructive bg-destructive text-destructive-foreground"}},defaultVariants:{variant:"default"}}),pr=i.forwardRef(({className:e,variant:t,...a},o)=>(Me(),n.jsx(Jo,{ref:o,className:se(Xs({variant:t}),e),...a})));pr.displayName=Jo.displayName;const el=i.forwardRef(({className:e,...t},a)=>
97n.jsx(tr,{ref:a,className:se("inline-flex h-8 shrink-0 items-center justify-center rounded-md border bg-transparent px-3 text-sm font-medium ring-offset-background transition-colors group-[.destructive]:border-muted/40 hover:bg-secondary group-[.destructive]:hover:border-destructive/30 group-[.destructive]:hover:bg-destructive group-[.destructive]:hover:text-destructive-foreground focus:outline-none focus:ring-2 focus:ring-ring focus:ring-offset-2 group-[.destructive]:focus:ring-destructive disabled:pointer-events-none disabled:opacity-50",e),...t}));el.displayName=tr.displayName;const fr=i.forwardRef(({className:e,...t},a)=>n.jsx(nr,{ref:a,className:se("absolute right-2 top-2 rounded-md p-1 text-foreground/50 opacity-0 transition-opacity group-hover:opacity-100 group-[.destructive]:text-red-300 hover:text-foreground group-[.destructive]:hover:text-red-50 focus:opacity-100 focus:outline-none focus:ring-2 group-[.destructive]:focus:ring-red-400 group-[.destructive]:focus:ring-offset-red-600",e),"toast-close":"",...t,children:n.jsx(xn,{className:"h-4 w-4"})}));fr.displayName=nr.displayName;const hr=i.forwardRef(({className:e,...t},a)=>n.jsx(Xo,{ref:a,className:se("text-sm font-semibold",e),...t}));hr.displayName=Xo.displayName;const gr=i.forwardRef(({className:e,...t},a)=>n.jsx(er,{ref:a,className:se("text-sm opacity-90",e),...t}));gr.displayName=er.displayName;function tl(){const{toasts:e}=Gi();return n.jsxs(Js,{children:[e.map(function({id:t,title:a,description:o,action:r,...s}){return n.jsxs(pr,{...s,children:[n.jsxs("div",{className:"grid gap-1",children:[a&&n.jsx(hr,{children:a}),o&&n.jsx(gr,{children:o})]}),r,n.jsx(fr,{})]},t)}),n.jsx(mr,{})]})}var yn=["light","dark"],fa="(prefers-color-scheme: dark)",nl=typeof window>"u",ha=i.createContext(void 0),al={setTheme:e=>{},themes:[]},ol=()=>{var e;return(e=i.useContext(ha))!=null?e:al},rl=e=>i.useContext(ha)?e.children:i.createElement(sl,{...e}),il=["light","dark"],sl=({forcedTheme:e,disableTransitionOnChange:t=!1,enableSystem:a=!0,enableColorScheme:o=!0,storageKey:r="theme",themes:s=il,defaultTheme:d=a?"system":"light",attribute:l="data-theme",value:m,children:u,nonce:v})=>{let[h,g]=i.useState(()=>_a(r,d)),[c,w]=i.useState(()=>_a(r)),y=m?Object.values(m):s,b=i.useCallback(P=>{let T=P;if(!T)return;P==="system"&&a&&(T=za());let E=m?m[T]:T,j=t?cl():null,D=document.documentElement;if(l==="class"?(D.classList.remove(...y),E&&D.classList.add(E)):E?D.setAttribute(l,E):D.removeAttribute(l),o){let q=yn.includes(d)?d:null,B=yn.includes(T)?T:q;D.style.colorScheme=B}j==null||j()},[]),x=i.useCallback(P=>{let T=typeof P=="function"?P(P):P;g(T);try{localStorage.setItem(r,T)}catch{}},[e]),I=i.useCallback(P=>{let T=za(P);w(T),h==="system"&&a&&!e&&b("system")},[h,e]);i.useEffect(()=>{let P=window.matchMedia(fa);return P.addListener(I),I(P),()=>P.removeListener(I)},[I]),i.useEffect(()=>{let P=T=>{if(T.key!==r)return;let E=T.newValue||d;x(E)};return window.addEventListener("storage",P),()=>window.removeEventListener("storage",P)},[x]),i.useEffect(()=>{b(e??h)},[e,h]);let k=i.useMemo(()=>({theme:h,setTheme:x,forcedTheme:e,resolvedTheme:h==="system"?c:h,themes:a?[...s,"system"]:s,systemTheme:a?c:void 0}),[h,x,e,c,a,s]);return i.createElement(ha.Provider,{value:k},i.createElement(ll,{forcedTheme:e,disableTransitionOnChange:t,enableSystem:a,enableColorScheme:o,storageKey:r,themes:s,defaultTheme:d,attribute:l,value:m,children:u,attrs:y,nonce:v}),u)},ll=i.memo(({forcedTheme:e,storageKey:t,attribute:a,enableSystem:o,enableColorScheme:r,defaultTheme:s,value:d,attrs:l,nonce:m})=>{let u=s==="system",v=a==="class"?`var d=document.documentElement,c=d.classList;${`c.remove(${l.map(w=>`'${w}'`).join(",")})`};`:`var d=document.documentElement,n='${a}',s='setAttribute';`,h=r?yn.includes(s)&&s?`if(e==='light'||e==='dark'||!e)d.style.colorScheme=e||'${s}'`:"if(e==='light'||e==='dark')d.style.colorScheme=e":"",g=(w,y=!1,b=!0)=>{let x=d?d[w]:w,I=y?w+"|| ''":`'${x}'`,k="";return r&&b&&!y&&yn.includes(w)&&(k+=`d.style.colorScheme = '${w}';`),a==="class"?y||x?k+=`c.add(${I})`:k+="null":x&&(k+=`d[s](n,${I})`),k},c=e?`!function(){${v}${g(e)}}()`:o?`!function(){try{${v}var e=localStorage.getItem('${t}');if('system'===e||(!e&&${u})){var t='${fa}',m=window.matchMedia(t);if(m.media!==t||m.matches){${g("dark")}}else{${g("light")}}}else if(e){${d?`var x=${JSON.stringify(d)};`:""}${g(d?"x[e]":"e",!0)}}${u?"":"else{"+g(s,!1,!1)+"}"}${h}}catch(e){}}()`:`!function(){try{${v}var e=localStorage.getItem('${t}');if(e){${d?`var x=${JSON.stringify(d)};`:""}${g(d?"x[e]":"e",!0)}}else{${g(s,!1,!1)};}${h}}catch(t){}}();`;return i.createElement("script",{nonce:m,dangerouslySetInnerHTML:{__html:c}})}),_a=(e,t)=>{if(nl)return;let a;try{a=localStorage.getItem(e)||void 0}catch{}return a||t},cl=()=>{let e=document.createElement("style");return e.appendChild(document.createTextNode("*{-webkit-transition:none!important;-moz-transition:none!important;-o-transition:none!important;-ms-transition:none!important;transition:none!important}")),document.head.appendChild(e),()=>{window.getComputedStyle(document.body),setTimeout(()=>{document.head.removeChild(e)},1)}},za=e=>
97(e||(e=window.matchMedia(fa)),e.matches?"dark":"light"),dl=e=>{switch(e){case"success":return pl;case"info":return hl;case"warning":return fl;case"error":return gl;default:return null}},ul=Array(12).fill(0),ml=({visible:e,className:t})=>C.createElement("div",{className:["sonner-loading-wrapper",t].filter(Boolean).join(" "),"data-visible":e},C.createElement("div",{className:"sonner-spinner"},ul.map((a,o)=>C.createElement("div",{className:"sonner-loading-bar",key:`spinner-bar-${o}`})))),pl=C.createElement("svg",{xmlns:"http://www.w3.org/2000/svg",viewBox:"0 0 20 20",fill:"currentColor",height:"20",width:"20"},C.createElement("path",{fillRule:"evenodd",d:"M10 18a8 8 0 100-16 8 8 0 000 16zm3.857-9.809a.75.75 0 00-1.214-.882l-3.483 4.79-1.88-1.88a.75.75 0 10-1.06 1.061l2.5 2.5a.75.75 0 001.137-.089l4-5.5z",clipRule:"evenodd"})),fl=C.createElement("svg",{xmlns:"http://www.w3.org/2000/svg",viewBox:"0 0 24 24",fill:"currentColor",height:"20",width:"20"},C.createElement("path",{fillRule:"evenodd",d:"M9.401 3.003c1.155-2 4.043-2 5.197 0l7.355 12.748c1.154 2-.29 4.5-2.599 4.5H4.645c-2.309 0-3.752-2.5-2.598-4.5L9.4 3.003zM12 8.25a.75.75 0 01.75.75v3.75a.75.75 0 01-1.5 0V9a.75.75 0 01.75-.75zm0 8.25a.75.75 0 100-1.5.75.75 0 000 1.5z",clipRule:"evenodd"})),hl=C.createElement("svg",{xmlns:"http://www.w3.org/2000/svg",viewBox:"0 0 20 20",fill:"currentColor",height:"20",width:"20"},C.createElement("path",{fillRule:"evenodd",d:"M18 10a8 8 0 11-16 0 8 8 0 0116 0zm-7-4a1 1 0 11-2 0 1 1 0 012 0zM9 9a.75.75 0 000 1.5h.253a.25.25 0 01.244.304l-.459 2.066A1.75 1.75 0 0010.747 15H11a.75.75 0 000-1.5h-.253a.25.25 0 01-.244-.304l.459-2.066A1.75 1.75 0 009.253 9H9z",clipRule:"evenodd"})),gl=C.createElement("svg",{xmlns:"http://www.w3.org/2000/svg",viewBox:"0 0 20 20",fill:"currentColor",height:"20",width:"20"},C.createElement("path",{fillRule:"evenodd",d:"M18 10a8 8 0 11-16 0 8 8 0 0116 0zm-8-5a.75.75 0 01.75.75v4.5a.75.75 0 01-1.5 0v-4.5A.75.75 0 0110 5zm0 10a1 1 0 100-2 1 1 0 000 2z",clipRule:"evenodd"})),bl=C.createElement("svg",{xmlns:"http://www.w3.org/2000/svg",width:"12",height:"12",viewBox:"0 0 24 24",fill:"none",stroke:"currentColor",strokeWidth:"1.5",strokeLinecap:"round",strokeLinejoin:"round"},C.createElement("line",{x1:"18",y1:"6",x2:"6",y2:"18"}),C.createElement("line",{x1:"6",y1:"6",x2:"18",y2:"18"})),yl=()=>{let[e,t]=C.useState(document.hidden);return C.useEffect(()=>{let a=()=>{t(document.hidden)};return document.addEventListener("visibilitychange",a),()=>window.removeEventListener("visibilitychange",a)},[]),e},Jn=1,vl=class{constructor(){this.subscribe=e=>(this.subscribers.push(e),()=>{let t=this.subscribers.indexOf(e);this.subscribers.splice(t,1)}),this.publish=e=>{this.subscribers.forEach(t=>t(e))},this.addToast=e=>{this.publish(e),this.toasts=[...this.toasts,e]},this.create=e=>{var t;let{message:a,...o}=e,r=typeof(e==null?void 0:e.id)=="number"||((t=e.id)==null?void 0:t.length)>0?e.id:Jn++,s=this.toasts.find(l=>l.id===r),d=e.dismissible===void 0?!0:e.dismissible;return this.dismissedToasts.has(r)&&this.dismissedToasts.delete(r),s?this.toasts=this.toasts.map(l=>l.id===r?(this.publish({...l,...e,id:r,title:a}),{...l,...e,id:r,dismissible:d,title:a}):l):this.addToast({title:a,...o,dismissible:d,id:r}),r},this.dismiss=e=>(this.dismissedToasts.add(e),e||this.toasts.forEach(t=>{this.subscribers.forEach(a=>a({id:t.id,dismiss:!0}))}),this.subscribers.forEach(t=>t({id:e,dismiss:!0})),e),this.message=(e,t)=>this.create({...t,message:e}),this.error=(e,t)=>this.create({...t,message:e,type:"error"}),this.success=(e,t)=>this.create({...t,type:"success",message:e}),this.info=(e,t)=>this.create({...t,type:"info",message:e}),this.warning=(e,t)=>this.create({...t,type:"warning",message:e}),this.loading=(e,t)=>
97this.create({...t,type:"loading",message:e}),this.promise=(e,t)=>{if(!t)return;let a;t.loading!==void 0&&(a=this.create({...t,promise:e,type:"loading",message:t.loading,description:typeof t.description!="function"?t.description:void 0}));let o=e instanceof Promise?e:e(),r=a!==void 0,s,d=o.then(async m=>{if(s=["resolve",m],C.isValidElement(m))r=!1,this.create({id:a,type:"default",message:m});else if(Al(m)&&!m.ok){r=!1;let u=typeof t.error=="function"?await t.error(`HTTP error! status: ${m.status}`):t.error,v=typeof t.description=="function"?await t.description(`HTTP error! status: ${m.status}`):t.description;this.create({id:a,type:"error",message:u,description:v})}else if(t.success!==void 0){r=!1;let u=typeof t.success=="function"?await t.success(m):t.success,v=typeof t.description=="function"?await t.description(m):t.description;this.create({id:a,type:"success",message:u,description:v})}}).catch(async m=>{if(s=["reject",m],t.error!==void 0){r=!1;let u=typeof t.error=="function"?await t.error(m):t.error,v=typeof t.description=="function"?await t.description(m):t.description;this.create({id:a,type:"error",message:u,description:v})}}).finally(()=>{var m;r&&(this.dismiss(a),a=void 0),(m=t.finally)==null||m.call(t)}),l=()=>new Promise((m,u)=>d.then(()=>s[0]==="reject"?u(s[1]):m(s[1])).catch(u));return typeof a!="string"&&typeof a!="number"?{unwrap:l}:Object.assign(a,{unwrap:l})},this.custom=(e,t)=>{let a=(t==null?void 0:t.id)||Jn++;return this.create({jsx:e(a),id:a,...t}),a},this.getActiveToasts=()=>this.toasts.filter(e=>!this.dismissedToasts.has(e.id)),this.subscribers=[],this.toasts=[],this.dismissedToasts=new Set}},be=new vl,Il=(e,t)=>{let a=(t==null?void 0:t.id)||Jn++;return be.addToast({title:e,...t,id:a}),a},Al=e=>e&&typeof e=="object"&&"ok"in e&&typeof e.ok=="boolean"&&"status"in e&&typeof e.status=="number",xl=Il,wl=()=>be.toasts,Sl=()=>be.getActiveToasts(),Pg=Object.assign(xl,{success:be.success,info:be.info,warning:be.warning,error:be.error,custom:be.custom,message:be.message,promise:be.promise,dismiss:be.dismiss,loading:be.loading},{getHistory:wl,getToasts:Sl});function kl(e,{insertAt:t}={}){if(typeof document>"u")return;let a=document.head||document.getElementsByTagName("head")[0],o=document.createElement("style");o.type="text/css",t==="top"&&a.firstChild?a.insertBefore(o,a.firstChild):a.appendChild(o),o.styleSheet?o.styleSheet.cssText=e:o.appendChild(document.createTextNode(e))}kl(`:where(html[dir="ltr"]),:where([data-sonner-toaster][dir="ltr"]){--toast-icon-margin-start: -3px;--toast-icon-margin-end: 4px;--toast-svg-margin-start: -1px;--toast-svg-margin-end: 0px;--toast-button-margin-start: auto;--toast-button-margin-end: 0;--toast-close-button-start: 0;--toast-close-button-end: unset;--toast-close-button-transform: translate(-35%, -35%)}:where(html[dir="rtl"]),:where([data-sonner-toaster][dir="rtl"]){--toast-icon-margin-start: 4px;--toast-icon-margin-end: -3px;--toast-svg-margin-start: 0px;--toast-svg-margin-end: -1px;--toast-button-margin-start: 0;--toast-button-margin-end: auto;--toast-close-button-start: unset;--toast-close-button-end: 0;--toast-close-button-transform: translate(35%, -35%)}:where([data-sonner-toaster]){position:fixed;width:var(--width);font-family:ui-sans-serif,system-ui,-apple-system,BlinkMacSystemFont,Segoe UI,Roboto,Helvetica Neue,Arial,Noto Sans,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol,Noto Color Emoji;--gray1: hsl(0, 0%, 99%);--gray2: hsl(0, 0%, 97.3%);--gray3: hsl(0, 0%, 95.1%);--gray4: hsl(0, 0%, 93%);--gray5: hsl(0, 0%, 90.9%);--gray6: hsl(0, 0%, 88.7%);--gray7: hsl(0, 0%, 85.8%);--gray8: hsl(0, 0%, 78%);--gray9: hsl(0, 0%, 56.1%);--gray10: hsl(0, 0%, 52.3%);--gray11: hsl(0, 0%, 43.5%);--gray12: hsl(0, 0%, 9%);--border-radius: 8px;box-sizing:border-box;padding:0;margin:0;list-style:none;outline:none;z-index:999999999;transition:transform .4s ease}:where([data-sonner-toaster][data-lifted="true"]){transform:translateY(-10px)}@media (hover: none) and (pointer: coarse){:where([data-sonner-toaster][data-lifted="true"]){transform:none}}:where([data-sonner-toaster][data-x-position="right"]){right:var(--offset-right)}:where([data-sonner-toaster][data-x-position="left"]){left:var(--offset-left)}:where([data-sonner-toaster][data-x-position="center"]){left:50%;transform:translate(-50%)}:where([data-sonner-toaster][data-y-position="top"]){top:var(--offset-top)}:where([data-sonner-toaster][data-y-position="bottom"]){bottom:var(--offset-bottom)}:where([data-sonner-toast]){--y: translateY(100%);--lift-amount: calc(var(--lift) * var(--gap));z-index:var(--z-index);position:absolute;opacity:0;transform:var(--y);filter:blur(0);touch-action:none;transition:transform .4s,opacity .4s,height .4s,box-shadow .2s;box-sizing:border-box;outline:none;overflow-wrap:anywhere}:where([data-sonner-toast][data-styled="true"]){padding:16px;background:var(--normal-bg);border:1px solid var(--normal-border);color:var(--normal-text);border-radius:var(--border-radius);box-shadow:0 4px 12px #0000001a;width:var(--width);font-size:13px;display:flex;align-items:center;gap:6px}:where([data-sonner-toast]:focus-visible){box-shadow:0 4px 12px #0000001a,0 0 0 2px #0003}:where([data-sonner-toast][data-y-position="top"]){top:0;--y: translateY(-100%);--lift: 1;--lift-amount: calc(1 * var(--gap))}:where([data-sonner-toast][data-y-position="bottom"]){bottom:0;--y: translateY(100%);--lift: -1;--lift-amount: calc(var(--lift) * var(--gap))}:where([data-sonner-toast]) :where([data-description]){font-weight:400;line-height:1.4;color:inherit}:where([data-sonner-toast]) :where([data-title]){font-weight:500;line-height:1.5;color:inherit}:where([data-sonner-toast]) :where([data-icon]){display:flex;height:16px;width:16px;position:relative;justify-content:flex-start;align-items:center;flex-shrink:0;margin-left:var(--toast-icon-margin-start);margin-right:var(--toast-icon-margin-end)}:where([data-sonner-toast][data-promise="true"]) :where([data-icon])>svg{opacity:0;transform:scale(.8);transform-origin:center;animation:sonner-fade-in .3s ease forwards}:where([data-sonner-toast]) :where([data-icon])>*{flex-shrink:0}:where([data-sonner-toast]) :where([data-icon]) svg{margin-left:var(--toast-svg-margin-start);margin-right:var(--toast-svg-margin-end)}:where([data-sonner-toast]) :where([data-content]){display:flex;flex-direction:column;gap:2px}[data-sonner-toast][data-styled=true] [data-button]{border-radius:4px;padding-left:8px;padding-right:8px;height:24px;font-size:12px;color:var(--normal-bg);background:var(--normal-text);margin-left:var(--toast-button-margin-start);margin-right:var(--toast-button-margin-end);border:none;cursor:pointer;outline:none;display:flex;align-items:center;flex-shrink:0;transition:opacity .4s,box-shadow .2s}:where([data-sonner-toast]) :where([data-button]):focus-visible{box-shadow:0 0 0 2px #0006}:where([data-sonner-toast]) :where([data-button]):first-of-type{margin-left:var(--toast-button-margin-start);margin-right:var(--toast-button-margin-end)}:where([data-sonner-toast]) :where([data-cancel]){color:var(--normal-text);background:rgba(0,0,0,.08)}:where([data-sonner-toast][data-theme="dark"]) :where([data-cancel]){background:rgba(255,255,255,.3)}:where([data-sonner-toast]) :where([data-close-button]){position:absolute;
97left:var(--toast-close-button-start);right:var(--toast-close-button-end);top:0;height:20px;width:20px;display:flex;justify-content:center;align-items:center;padding:0;color:var(--gray12);border:1px solid var(--gray4);transform:var(--toast-close-button-transform);border-radius:50%;cursor:pointer;z-index:1;transition:opacity .1s,background .2s,border-color .2s}[data-sonner-toast] [data-close-button]{background:var(--gray1)}:where([data-sonner-toast]) :where([data-close-button]):focus-visible{box-shadow:0 4px 12px #0000001a,0 0 0 2px #0003}:where([data-sonner-toast]) :where([data-disabled="true"]){cursor:not-allowed}:where([data-sonner-toast]):hover :where([data-close-button]):hover{background:var(--gray2);border-color:var(--gray5)}:where([data-sonner-toast][data-swiping="true"]):before{content:"";position:absolute;left:-50%;right:-50%;height:100%;z-index:-1}:where([data-sonner-toast][data-y-position="top"][data-swiping="true"]):before{bottom:50%;transform:scaleY(3) translateY(50%)}:where([data-sonner-toast][data-y-position="bottom"][data-swiping="true"]):before{top:50%;transform:scaleY(3) translateY(-50%)}:where([data-sonner-toast][data-swiping="false"][data-removed="true"]):before{content:"";position:absolute;inset:0;transform:scaleY(2)}:where([data-sonner-toast]):after{content:"";position:absolute;left:0;height:calc(var(--gap) + 1px);bottom:100%;width:100%}:where([data-sonner-toast][data-mounted="true"]){--y: translateY(0);opacity:1}:where([data-sonner-toast][data-expanded="false"][data-front="false"]){--scale: var(--toasts-before) * .05 + 1;--y: translateY(calc(var(--lift-amount) * var(--toasts-before))) scale(calc(-1 * var(--scale)));height:var(--front-toast-height)}:where([data-sonner-toast])>*{transition:opacity .4s}:where([data-sonner-toast][data-expanded="false"][data-front="false"][data-styled="true"])>*{opacity:0}
97:where([data-sonner-toast][data-visible="false"]){opacity:0;pointer-events:none}:where([data-sonner-toast][data-mounted="true"][data-expanded="true"]){--y: translateY(calc(var(--lift) * var(--offset)));height:var(--initial-height)}:where([data-sonner-toast][data-removed="true"][data-front="true"][data-swipe-out="false"]){--y: translateY(calc(var(--lift) * -100%));opacity:0}:where([data-sonner-toast][data-removed="true"][data-front="false"][data-swipe-out="false"][data-expanded="true"]){--y: translateY(calc(var(--lift) * var(--offset) + var(--lift) * -100%));opacity:0}:where([data-sonner-toast][data-removed="true"][data-front="false"][data-swipe-out="false"][data-expanded="false"]){--y: translateY(40%);opacity:0;transition:transform .5s,opacity .2s}:where([data-sonner-toast][data-removed="true"][data-front="false"]):before{height:calc(var(--initial-height) + 20%)}[data-sonner-toast][data-swiping=true]{transform:var(--y) translateY(var(--swipe-amount-y, 0px)) translate(var(--swipe-amount-x, 0px));transition:none}[data-sonner-toast][data-swiped=true]{user-select:none}[data-sonner-toast][data-swipe-out=true][data-y-position=bottom],[data-sonner-toast][data-swipe-out=true][data-y-position=top]{animation-duration:.2s;animation-timing-function:ease-out;animation-fill-mode:forwards}[data-sonner-toast][data-swipe-out=true][data-swipe-direction=left]{animation-name:swipe-out-left}[data-sonner-toast][data-swipe-out=true][data-swipe-direction=right]{animation-name:swipe-out-right}[data-sonner-toast][data-swipe-out=true][data-swipe-direction=up]{animation-name:swipe-out-up}[data-sonner-toast][data-swipe-out=true][data-swipe-direction=down]{animation-name:swipe-out-down}@keyframes swipe-out-left{0%{transform:var(--y) translate(var(--swipe-amount-x));opacity:1}to{transform:var(--y) translate(calc(var(--swipe-amount-x) - 100%));opacity:0}}@keyframes swipe-out-right{0%{transform:var(--y) translate(var(--swipe-amount-x));opacity:1}to{transform:var(--y) translate(calc(var(--swipe-amount-x) + 100%));opacity:0}}@keyframes swipe-out-up{0%{transform:var(--y) translateY(var(--swipe-amount-y));opacity:1}to{transform:var(--y) translateY(calc(var(--swipe-amount-y) - 100%));opacity:0}}@keyframes swipe-out-down{0%{transform:var(--y) translateY(var(--swipe-amount-y));opacity:1}to{transform:var(--y) translateY(calc(var(--swipe-amount-y) + 100%));opacity:0}}@media (max-width: 600px){[data-sonner-toaster]{position:fixed;right:var(--mobile-offset-right);left:var(--mobile-offset-left);width:100%}[data-sonner-toaster][dir=rtl]{left:calc(var(--mobile-offset-left) * -1)}[data-sonner-toaster] [data-sonner-toast]{left:0;right:0;width:calc(100% - var(--mobile-offset-left) * 2)}[data-sonner-toaster][data-x-position=left]{left:var(--mobile-offset-left)}[data-sonner-toaster][data-y-position=bottom]{bottom:var(--mobile-offset-bottom)}[data-sonner-toaster][data-y-position=top]{top:var(--mobile-offset-top)}[data-sonner-toaster][data-x-position=center]{left:var(--mobile-offset-left);right:var(--mobile-offset-right);transform:none}}[data-sonner-toaster][data-theme=light]{--normal-bg: #fff;--normal-border: var(--gray4);--normal-text: var(--gray12);--success-bg: hsl(143, 85%, 96%);--success-border: hsl(145, 92%, 91%);--success-text: hsl(140, 100%, 27%);--info-bg: hsl(208, 100%, 97%);--info-border: hsl(221, 91%, 91%);--info-text: hsl(210, 92%, 45%);--warning-bg: hsl(49, 100%, 97%);--warning-border: hsl(49, 91%, 91%);--warning-text: hsl(31, 92%, 45%);--error-bg: hsl(359, 100%, 97%);--error-border: hsl(359, 100%, 94%);--error-text: hsl(360, 100%, 45%)}[data-sonner-toaster][data-theme=light] [data-sonner-toast][data-invert=true]{--normal-bg: #000;--normal-border: hsl(0, 0%, 20%);--normal-text: var(--gray1)}[data-sonner-toaster][data-theme=dark] [data-sonner-toast][data-invert=true]{--normal-bg: #fff;--normal-border: var(--gray3);--normal-text: var(--gray12)}[data-sonner-toaster][data-theme=dark]{--normal-bg: #000;--normal-bg-hover: hsl(0, 0%, 12%);--normal-border: hsl(0, 0%, 20%);--normal-border-hover: hsl(0, 0%, 25%);--normal-text: var(--gray1);--success-bg: hsl(150, 100%, 6%);--success-border: hsl(147, 100%, 12%);--success-text: hsl(150, 86%, 65%);--info-bg: hsl(215, 100%, 6%);--info-border: hsl(223, 100%, 12%);--info-text: hsl(216, 87%, 65%);--warning-bg: hsl(64, 100%, 6%);--warning-border: hsl(60, 100%, 12%);--warning-text: hsl(46, 87%, 65%);--error-bg: hsl(358, 76%, 10%);--error-border: hsl(357, 89%, 16%);
97--error-text: hsl(358, 100%, 81%)}[data-sonner-toaster][data-theme=dark] [data-sonner-toast] [data-close-button]{background:var(--normal-bg);border-color:var(--normal-border);color:var(--normal-text)}[data-sonner-toaster][data-theme=dark] [data-sonner-toast] [data-close-button]:hover{background:var(--normal-bg-hover);border-color:var(--normal-border-hover)}[data-rich-colors=true][data-sonner-toast][data-type=success],[data-rich-colors=true][data-sonner-toast][data-type=success] [data-close-button]{background:var(--success-bg);border-color:var(--success-border);color:var(--success-text)}[data-rich-colors=true][data-sonner-toast][data-type=info],[data-rich-colors=true][data-sonner-toast][data-type=info] [data-close-button]{background:var(--info-bg);border-color:var(--info-border);color:var(--info-text)}[data-rich-colors=true][data-sonner-toast][data-type=warning],[data-rich-colors=true][data-sonner-toast][data-type=warning] [data-close-button]{background:var(--warning-bg);border-color:var(--warning-border);color:var(--warning-text)}[data-rich-colors=true][data-sonner-toast][data-type=error],[data-rich-colors=true][data-sonner-toast][data-type=error] [data-close-button]{background:var(--error-bg);border-color:var(--error-border);color:var(--error-text)}.sonner-loading-wrapper{--size: 16px;height:var(--size);width:var(--size);position:absolute;inset:0;z-index:10}.sonner-loading-wrapper[data-visible=false]{transform-origin:center;animation:sonner-fade-out .2s ease forwards}.sonner-spinner{position:relative;top:50%;left:50%;height:var(--size);width:var(--size)}.sonner-loading-bar{animation:sonner-spin 1.2s linear infinite;background:var(--gray11);border-radius:6px;height:8%;left:-10%;position:absolute;top:-3.9%;width:24%}.sonner-loading-bar:nth-child(1){animation-delay:-1.2s;transform:rotate(.0001deg) translate(146%)}.sonner-loading-bar:nth-child(2){animation-delay:-1.1s;transform:rotate(30deg) translate(146%)}.sonner-loading-bar:nth-child(3){animation-delay:-1s;transform:rotate(60deg) translate(146%)}.sonner-loading-bar:nth-child(4){animation-delay:-.9s;transform:rotate(90deg) translate(146%)}.sonner-loading-bar:nth-child(5){animation-delay:-.8s;transform:rotate(120deg) translate(146%)}.sonner-loading-bar:nth-child(6){animation-delay:-.7s;transform:rotate(150deg) translate(146%)}.sonner-loading-bar:nth-child(7){animation-delay:-.6s;transform:rotate(180deg) translate(146%)}.sonner-loading-bar:nth-child(8){animation-delay:-.5s;transform:rotate(210deg) translate(146%)}.sonner-loading-bar:nth-child(9){animation-delay:-.4s;transform:rotate(240deg) translate(146%)}.sonner-loading-bar:nth-child(10){animation-delay:-.3s;transform:rotate(270deg) translate(146%)}.sonner-loading-bar:nth-child(11){animation-delay:-.2s;transform:rotate(300deg) translate(146%)}.sonner-loading-bar:nth-child(12){animation-delay:-.1s;transform:rotate(330deg) translate(146%)}@keyframes sonner-fade-in{0%{opacity:0;transform:scale(.8)}to{opacity:1;transform:scale(1)}}@keyframes sonner-fade-out{0%{opacity:1;transform:scale(1)}to{opacity:0;transform:scale(.8)}}@keyframes sonner-spin{0%{opacity:1}to{opacity:.15}}@media (prefers-reduced-motion){[data-sonner-toast],[data-sonner-toast]>*,.sonner-loading-bar{transition:none!important;animation:none!important}}.sonner-loader{position:absolute;top:50%;left:50%;transform:translate(-50%,-50%);transform-origin:center;transition:opacity .2s,transform .2s}.sonner-loader[data-visible=false]{opacity:0;transform:scale(.8) translate(-50%,-50%)} 98`);function on(e){return e.label!==void 0}var Cl=3,Pl="32px",Tl="16px",ja=4e3,El=356,Rl=14,Ol=20,Ml=200;function ke(...e){return e.filter(Boolean).join(" ")}function ql(e){let[t,a]=e.split("-"),o=[];return t&&o.push(t),a&&o.push(a),o}var Dl=e=>{var t,a,o,r,s,d,l,m,u,v,h;let{invert:g,toast:c,unstyled:w,interacting:y,setHeights:b,visibleToasts:x,heights:I,index:k,toasts:P,expanded:T,removeToast:E,defaultRichColors:j,closeButton:D,style:q,cancelButtonStyle:B,actionButtonStyle:G,className:S="",descriptionClassName:M="",duration:R,position:Y,gap:W,loadingIcon:ee,expandByDefault:le,classNames:O,icons:ne,closeButtonAriaLabel:ce="Close toast",pauseWhenPageIsHidden:
98Q}=e,[V,K]=C.useState(null),[fe,bt]=C.useState(null),[J,Cn]=C.useState(!1),[_t,en]=C.useState(!1),[zt,Pn]=C.useState(!1),[va,Hr]=C.useState(!1),[Wr,Ia]=C.useState(!1),[Kr,Tn]=C.useState(0),[$r,Aa]=C.useState(0),jt=C.useRef(c.duration||R||ja),xa=C.useRef(null),rt=C.useRef(null),Yr=k===0,Zr=k+1<=x,ve=c.type,yt=c.dismissible!==!1,Jr=c.className||"",Xr=c.descriptionClassName||"",tn=C.useMemo(()=>I.findIndex(L=>L.toastId===c.id)||0,[I,c.id]),ei=C.useMemo(()=>{var L;return(L=c.closeButton)!=null?L:D},[c.closeButton,D]),wa=C.useMemo(()=>c.duration||R||ja,[c.duration,R]),En=C.useRef(0),vt=C.useRef(0),Sa=C.useRef(0),It=C.useRef(null),[ti,ni]=Y.split("-"),ka=C.useMemo(()=>I.reduce((L,H,Z)=>Z>=tn?L:L+H.height,0),[I,tn]),Ca=yl(),ai=c.invert||g,Rn=ve==="loading";vt.current=C.useMemo(()=>tn*W+ka,[tn,ka]),C.useEffect(()=>{jt.current=wa},[wa]),C.useEffect(()=>{Cn(!0)},[]),C.useEffect(()=>{let L=rt.current;if(L){let H=L.getBoundingClientRect().height;return Aa(H),b(Z=>[{toastId:c.id,height:H,position:c.position},...Z]),()=>b(Z=>Z.filter(xe=>xe.toastId!==c.id))}},[b,c.id]),C.useLayoutEffect(()=>{if(!J)return;let L=rt.current,H=L.style.height;L.style.height="auto";let Z=L.getBoundingClientRect().height;L.style.height=H,Aa(Z),b(xe=>xe.find(we=>we.toastId===c.id)?xe.map(we=>we.toastId===c.id?{...we,height:Z}:we):[{toastId:c.id,height:Z,position:c.position},...xe])},[J,c.title,c.description,b,c.id]);let He=C.useCallback(()=>{en(!0),Tn(vt.current),b(L=>L.filter(H=>H.toastId!==c.id)),setTimeout(()=>{E(c)},Ml)},[c,E,b,vt]);C.useEffect(()=>{if(c.promise&&ve==="loading"||c.duration===1/0||c.type==="loading")return;let L;return T||y||Q&&Ca?(()=>{if(Sa.current<En.current){let H=new Date().getTime()-En.current;jt.current=jt.current-H}Sa.current=new Date().getTime()})():jt.current!==1/0&&(En.current=new Date().getTime(),L=setTimeout(()=>{var H;(H=c.onAutoClose)==null||H.call(c,c),He()},jt.current)),()=>clearTimeout(L)},[T,y,c,ve,Q,Ca,He]),C.useEffect(()=>{c.delete&&He()},[He,c.delete]);function oi(){var L,H,Z;return ne!=null&&ne.loading?C.createElement("div",{className:ke(O==null?void 0:O.loader,(L=c==null?void 0:c.classNames)==null?void 0:L.loader,"sonner-loader"),"data-visible":ve==="loading"},ne.loading):ee?C.createElement("div",{className:ke(O==null?void 0:O.loader,(H=c==null?void 0:c.classNames)==null?void 0:H.loader,"sonner-loader"),"data-visible":ve==="loading"},ee):C.createElement(ml,{className:ke(O==null?void 0:O.loader,(Z=c==null?void 0:c.classNames)==null?void 0:Z.loader),visible:ve==="loading"})}return C.createElement("li",{tabIndex:0,ref:rt,className:ke(S,Jr,O==null?void 0:O.toast,(t=c==null?void 0:c.classNames)==null?void 0:t.toast,O==null?void 0:O.default,O==null?void 0:O[ve],(a=c==null?void 0:c.classNames)==null?void 0:a[ve]),"data-sonner-toast":"","data-rich-colors":(o=c.richColors)!=null?o:j,"data-styled":!(c.jsx||c.unstyled||w),"data-mounted":J,"data-promise":!!c.promise,"data-swiped":Wr,"data-removed":_t,"data-visible":Zr,"data-y-position":ti,"data-x-position":ni,"data-index":k,"data-front":Yr,"data-swiping":zt,"data-dismissible":yt,"data-type":ve,"data-invert":ai,"data-swipe-out":va,"data-swipe-direction":fe,"data-expanded":!!(T||le&&J),style:{"--index":k,"--toasts-before":k,"--z-index":P.length-k,"--offset":`${_t?Kr:vt.current}px`,"--initial-height":le?"auto":`${$r}px`,...q,...c.style},onDragEnd:()=>{Pn(!1),K(null),It.current=null},onPointerDown:L=>{Rn||!yt||(xa.current=new Date,Tn(vt.current),L.target.setPointerCapture(L.pointerId),L.target.tagName!=="BUTTON"&&(Pn(!0),It.current={x:L.clientX,y:L.clientY}))},onPointerUp:()=>{var L,H,Z,xe;if(va||!yt)return;It.current=null;let we=Number(((L=rt.current)==null?void 0:L.style.getPropertyValue("--swipe-amount-x").replace("px",""))||0),We=Number(((H=rt.current)==null?void 0:H.style.getPropertyValue("--swipe-amount-y").replace("px",""))||0),it=new Date().getTime()-((Z=xa.current)==null?void 0:Z.getTime()),Se=V==="x"?we:We,Ke=Math.abs(Se)/it;if(Math.abs(Se)>=Ol||Ke>.11){Tn(vt.current),(xe=c.onDismiss)==null||xe.call(c,c),bt(V==="x"?we>0?"right":"left":We>0?"down":"up"),He(),Hr(!0),Ia(!1);return}Pn(!1),K(null)},onPointerMove:L=>{var H,Z,xe,we;if(!It.current||!yt||((H=window.getSelection())==null?void 0:H.toString().length)>0)return;let We=L.clientY-It.current.y,it=L.clientX-It.current.x,Se=(Z=e.swipeDirections)!=null?Z:ql(Y);!V&&(Math.abs(it)>1||Math.abs(We)>1)&&K(Math.abs(it)>Math.abs(We)?"x":"y");let Ke={x:0,y:0};V==="y"?(Se.includes("top")||Se.includes("bottom"))&&(Se.includes("top")&&We
98<0||Se.includes("bottom")&&We>0)&&(Ke.y=We):V==="x"&&(Se.includes("left")||Se.includes("right"))&&(Se.includes("left")&&it<0||Se.includes("right")&&it>0)&&(Ke.x=it),(Math.abs(Ke.x)>0||Math.abs(Ke.y)>0)&&Ia(!0),(xe=rt.current)==null||xe.style.setProperty("--swipe-amount-x",`${Ke.x}px`),(we=rt.current)==null||we.style.setProperty("--swipe-amount-y",`${Ke.y}px`)}},ei&&!c.jsx?C.createElement("button",{"aria-label":ce,"data-disabled":Rn,"data-close-button":!0,onClick:Rn||!yt?()=>{}:()=>{var L;He(),(L=c.onDismiss)==null||L.call(c,c)},className:ke(O==null?void 0:O.closeButton,(r=c==null?void 0:c.classNames)==null?void 0:r.closeButton)},(s=ne==null?void 0:ne.close)!=null?s:bl):null,c.jsx||i.isValidElement(c.title)?c.jsx?c.jsx:typeof c.title=="function"?c.title():c.title:C.createElement(C.Fragment,null,ve||c.icon||c.promise?C.createElement("div",{"data-icon":"",className:ke(O==null?void 0:O.icon,(d=c==null?void 0:c.classNames)==null?void 0:d.icon)},c.promise||c.type==="loading"&&!c.icon?c.icon||oi():null,c.type!=="loading"?c.icon||(ne==null?void 0:ne[ve])||dl(ve):null):null,C.createElement("div",{"data-content":"",className:ke(O==null?void 0:O.content,(l=c==null?void 0:c.classNames)==null?void 0:l.content)},C.createElement("div",{"data-title":"",className:ke(O==null?void 0:O.title,(m=c==null?void 0:c.classNames)==null?void 0:m.title)},typeof c.title=="function"?c.title():c.title),c.description?C.createElement("div",{"data-description":"",className:ke(M,Xr,O==null?void 0:O.description,(u=c==null?void 0:c.classNames)==null?void 0:u.description)},typeof c.description=="function"?c.description():c.description):null),i.isValidElement(c.cancel)?c.cancel:c.cancel&&on(c.cancel)?C.createElement("button",{"data-button":!0,"data-cancel":!0,style:c.cancelButtonStyle||B,onClick:L=>{var H,Z;on(c.cancel)&&yt&&((Z=(H=c.cancel).onClick)==null||Z.call(H,L),He())},className:ke(O==null?void 0:O.cancelButton,(v=c==null?void 0:c.classNames)==null?void 0:v.cancelButton)},c.cancel.label):null,i.isValidElement(c.action)?c.action:c.action&&on(c.action)?C.createElement("button",{"data-button":!0,"data-action":!0,style:c.actionButtonStyle||G,onClick:L=>{var H,Z;on(c.action)&&((Z=(H=c.action).onClick)==null||Z.call(H,L),!L.defaultPrevented&&He())},className:ke(O==null?void 0:O.actionButton,(h=c==null?void 0:c.classNames)==null?void 0:h.actionButton)},c.action.label):null))};function Ga(){if(typeof window>"u"||typeof document>"u")return"ltr";let e=document.documentElement.getAttribute("dir");return e==="auto"||!e?window.getComputedStyle(document.documentElement).direction:e}function Ll(e,t){let a={};return[e,t].forEach((o,r)=>{let s=r===1,d=s?"--mobile-offset":"--offset",l=s?Tl:Pl;function m(u){["top","right","bottom","left"].forEach(v=>{a[`${d}-${v}`]=typeof u=="number"?`${u}px`:u})}typeof o=="number"||typeof o=="string"?m(o):typeof o=="object"?["top","right","bottom","left"].forEach(u=>{o[u]===void 0?a[`${d}-${u}`]=l:a[`${d}-${u}`]=typeof o[u]=="number"?`${o[u]}px`:o[u]}):m(l)}),a}var Fl=i.forwardRef(function(e,t){let{invert:a,position:o="bottom-right",hotkey:r=["altKey","KeyT"],expand:s,closeButton:d,className:l,offset:m,mobileOffset:u,theme:v="light",richColors:h,duration:g,style:c,visibleToasts:w=Cl,toastOptions:y,dir:b=Ga(),gap:x=Rl,loadingIcon:I,icons:k,containerAriaLabel:P="Notifications",pauseWhenPageIsHidden:T}=e,[E,j]=C.useState([]),D=C.useMemo(()=>Array.from(new Set([o].concat(E.filter(Q=>Q.position).map(Q=>Q.position)))),[E,o]),[q,B]=C.useState([]),[G,S]=C.useState(!1),[M,R]=C.useState(!1),[Y,W]=C.useState(v!=="system"?v:typeof window<"u"&&window.matchMedia&&window.matchMedia("(prefers-color-scheme: dark)").matches?"dark":"light"),ee=C.useRef(null),le=r.join("+").replace(/Key/g,"").replace(/Digit/g,""),O=C.useRef(null),ne=C.useRef(!1),ce=C.useCallback(Q=>{j(V=>{var K;return(K=V.find(fe=>fe.id===Q.id))!=null&&K.delete||be.dismiss(Q.id),V.filter(({id:fe})=>fe!==Q.id)})},[]);return C.useEffect(()=>be.subscribe(Q=>{if(Q.dismiss){j(V=>V.map(K=>K.id===Q.id?{...K,delete:!0}:K));return}setTimeout(()=>{li.flushSync(()=>{j(V=>{let K=V.findIndex(fe=>fe.id===Q.id);return K!==-1?[...V.slice(0,K),{...V[K],...Q},...V.slice(K+1)]:[Q,...V]})})})}),[]),C.useEffect(()=>{if(v!=="system"){W(v);return}if(v==="system"&&(window.matchMedia&&window.matchMedia("(prefers-color-scheme: dark)").matches?W("dark"):W("light")),typeof window>"u")return;let Q=window.matchMedia("(prefers-color-scheme: dark)");try{Q.addEventListener("change",({matches:V})=>{W(V?"dark":"light")})}catch{Q.addListener(({matches:K})=>{try{W(K?"dark":"light")}catch(fe){console.error(fe)}})}},[v]),C.useEffect(()=>{E.length<=1&&S(!1)},[E]),C.useEffect(()=>{let Q=V=>{var K,fe;r.every(bt=>V[bt]||V.code===bt)&&(S(!0),(K=ee.current)==null||K.focus()),V.code==="Escape"&&(document.activeElement===ee.current||(fe=ee.current)!=null&&fe.contains(document.activeElement))&&S(!1)};return document.addEventListener("keydown",Q),()=>document.removeEventListener("keydown",Q)},[r]),C.useEffect(()=>{if(ee.current)return()=>{O.current&&(O.current.focus({preventScroll:!0}),O.current=null,ne.current=!1)}},[ee.current]),C.createElement("section",{ref:t,"aria-label":`${P} ${le}`,tabIndex:-1,"aria-live":"polite","aria-relevant":"additions text","aria-atomic":"false",suppressHydrationWarning:!0}
98,D.map((Q,V)=>{var K;let[fe,bt]=Q.split("-");return E.length?C.createElement("ol",{key:Q,dir:b==="auto"?Ga():b,tabIndex:-1,ref:ee,className:l,"data-sonner-toaster":!0,"data-theme":Y,"data-y-position":fe,"data-lifted":G&&E.length>1&&!s,"data-x-position":bt,style:{"--front-toast-height":`${((K=q[0])==null?void 0:K.height)||0}px`,"--width":`${El}px`,"--gap":`${x}px`,...c,...Ll(m,u)},onBlur:J=>{ne.current&&!J.currentTarget.contains(J.relatedTarget)&&(ne.current=!1,O.current&&(O.current.focus({preventScroll:!0}),O.current=null))},onFocus:J=>{J.target instanceof HTMLElement&&J.target.dataset.dismissible==="false"||ne.current||(ne.current=!0,O.current=J.relatedTarget)},onMouseEnter:()=>S(!0),onMouseMove:()=>S(!0),onMouseLeave:()=>{M||S(!1)},onDragEnd:()=>S(!1),onPointerDown:J=>{J.target instanceof HTMLElement&&J.target.dataset.dismissible==="false"||R(!0)},onPointerUp:()=>R(!1)},E.filter(J=>!J.position&&V===0||J.position===Q).map((J,Cn)=>{var _t,en;return C.createElement(Dl,{key:J.id,icons:k,index:Cn,toast:J,defaultRichColors:h,duration:(_t=y==null?void 0:y.duration)!=null?_t:g,className:y==null?void 0:y.className,descriptionClassName:y==null?void 0:y.descriptionClassName,invert:a,visibleToasts:w,closeButton:(en=y==null?void 0:y.closeButton)!=null?en:d,interacting:M,position:Q,style:y==null?void 0:y.style,unstyled:y==null?void 0:y.unstyled,classNames:y==null?void 0:y.classNames,cancelButtonStyle:y==null?void 0:y.cancelButtonStyle,actionButtonStyle:y==null?void 0:y.actionButtonStyle,removeToast:ce,toasts:E.filter(zt=>zt.position==J.position),heights:q.filter(zt=>zt.position==J.position),setHeights:B,expandByDefault:s,gap:x,loadingIcon:I,expanded:G,pauseWhenPageIsHidden:T,swipeDirections:e.swipeDirections})})):null}))});const br=i.forwardRef(({...e},t)=>{Me();const{theme:a="system"}=ol();return n.jsx(Fl,{ref:t,theme:a,className:"toaster group",toastOptions:{classNames:{toast:"group toast group-[.toaster]:bg-background group-[.toaster]:text-foreground group-[.toaster]:border-border group-[.toaster]:shadow-lg",description:"group-[.toast]:text-muted-foreground",actionButton:"group-[.toast]:bg-primary group-[.toast]:text-primary-foreground",cancelButton:"group-[.toast]:bg-muted group-[.toast]:text-muted-foreground"}},...e})});br.displayName="Toaster";const Nl=vi,Tg=Ii,Eg=Ai,_l=i.forwardRef(({className:e,sideOffset:t=4,...a},o)=>n.jsx(Oo,{ref:o,sideOffset:t,className:se("z-50 overflow-hidden rounded-md border bg-popover px-3 py-1.5 text-sm text-popover-foreground shadow-md animate-in fade-in-0 zoom-in-95 data-[state=closed]:animate-out data-[state=closed]:fade-out-0 data-[state=closed]:zoom-out-95 data-[side=bottom]:slide-in-from-top-2 data-[side=left]:slide-in-from-right-2 data-[side=right]:slide-in-from-left-2 data-[side=top]:slide-in-from-bottom-2",e),...a}));_l.displayName=Oo.displayName;var wn=class{constructor(){this.listeners=new Set,this.subscribe=this.subscribe.bind(this)}subscribe(e){return this.listeners.add(e),this.onSubscribe(),()=>{this.listeners.delete(e),this.onUnsubscribe()}}hasListeners(){return this.listeners.size>0}onSubscribe(){}onUnsubscribe(){}},Sn=typeof window>"u"||"Deno"in globalThis;function Ce(){}function zl(e,t){return typeof e=="function"?e(t):e}function jl(e){return typeof e=="number"&&e>=0&&e!==1/0}function Gl(e,t){return Math.max(e+(t||0)-Date.now(),0)}function Xn(e,t){return typeof e=="function"?e(t):e}function Bl(e,t){return typeof e=="function"?e(t):e}function Ba(e,t){const{type:a="all",exact:o,fetchStatus:r,predicate:s,queryKey:d,stale:l}=e;if(d){if(o){if(t.queryHash!==ga(d,t.options))return!1}else if(!$t(t.queryKey,d))return!1}if(a!=="all"){const m=t.isActive();if(a==="active"&&!m||a==="inactive"&&m)return!1}return!(typeof l=="boolean"&&t.isStale()!==l||r&&r!==t.state.fetchStatus||s&&!s(t))}function Ua(e,t){const{exact:a,status:o,predicate:r,mutationKey:s}=e;if(s){if(!t.options.mutationKey)return!1;if(a){if(Kt(t.options.mutationKey)!==Kt(s))return!1}else if(!$t(t.options.mutationKey,s))return!1}return!(o&&t.state.status!==o||r&&!r(t))}function ga(e,t){return((t==null?void 0:t.queryKeyHashFn)||Kt)(e)}function Kt(e){return JSON.stringify(e,(t,a)=>ea(a)?Object.keys(a).sort().reduce((o,r)=>(o[r]=a[r],o),{}):a)}function $t(e,t){return e===t?!0:typeof e!=typeof t?!1:e&&t&&typeof e=="object"&&typeof t=="object"?Object.keys(t).every(a=>$t(e[a],t[a])):!1}function yr(e,t){if(e===t)return e;const a=Qa(e)&&Qa(t);if(a||ea(e)&&ea(t)){const o=a?e:Object.keys(e),r=o.length,s=a?t:Object.keys(t),d=s.length,l=a?[]:{},m=new Set(o);let u=0;for(let v=0;v<d;v++){const h=a?v:s[v];
vendor: 17,058 bytes, line 98
98(!a&&m.has(h)||a)&&e[h]===void 0&&t[h]===void 0?(l[h]=void 0,u++):(l[h]=yr(e[h],t[h]),l[h]===e[h]&&e[h]!==void 0&&u++)}return r===d&&u===r?e:l}return t}function Rg(e,t){if(!t||Object.keys(e).length!==Object.keys(t).length)return!1;for(const a in e)if(e[a]!==t[a])return!1;return!0}function Qa(e){return Array.isArray(e)&&e.length===Object.keys(e).length}function ea(e){if(!Va(e))return!1;const t=e.constructor;if(t===void 0)return!0;const a=t.prototype;return!(!Va(a)||!a.hasOwnProperty("isPrototypeOf")||Object.getPrototypeOf(e)!==Object.prototype)}function Va(e){return Object.prototype.toString.call(e)==="[object Object]"}function Ul(e){return new Promise(t=>{setTimeout(t,e)})}function Ql(e,t,a){return typeof a.structuralSharing=="function"?a.structuralSharing(e,t):a.structuralSharing!==!1?yr(e,t):t}function Vl(e,t,a=0){const o=[...e,t];return a&&o.length>a?o.slice(1):o}function Hl(e,t,a=0){const o=[t,...e];return a&&o.length>a?o.slice(0,-1):o}var ba=Symbol();function vr(e,t){return!e.queryFn&&(t!=null&&t.initialPromise)?()=>t.initialPromise:!e.queryFn||e.queryFn===ba?()=>Promise.reject(new Error(`Missing queryFn: '${e.queryHash}'`)):e.queryFn}function Og(e,t){return typeof e=="function"?e(...t):!!e}var dt,Xe,Pt,yo,Wl=(yo=class extends wn{constructor(){super();U(this,dt);U(this,Xe);U(this,Pt);F(this,Pt,t=>{if(!Sn&&window.addEventListener){const a=()=>t();return window.addEventListener("visibilitychange",a,!1),()=>{window.removeEventListener("visibilitychange",a)}}})}onSubscribe(){A(this,Xe)||this.setEventListener(A(this,Pt))}onUnsubscribe(){var t;this.hasListeners()||((t=A(this,Xe))==null||t.call(this),F(this,Xe,void 0))}setEventListener(t){var a;F(this,Pt,t),(a=A(this,Xe))==null||a.call(this),F(this,Xe,t(o=>{typeof o=="boolean"?this.setFocused(o):this.onFocus()}))}setFocused(t){A(this,dt)!==t&&(F(this,dt,t),this.onFocus())}onFocus(){const t=this.isFocused();this.listeners.forEach(a=>{a(t)})}isFocused(){var t;return typeof A(this,dt)=="boolean"?A(this,dt):((t=globalThis.document)==null?void 0:t.visibilityState)!=="hidden"}},dt=new WeakMap,Xe=new WeakMap,Pt=new WeakMap,yo),Ir=new Wl,Tt,et,Et,vo,Kl=(vo=class extends wn{constructor(){super();U(this,Tt,!0);U(this,et);U(this,Et);F(this,Et,t=>{if(!Sn&&window.addEventListener){const a=()=>t(!0),o=()=>t(!1);return window.addEventListener("online",a,!1),window.addEventListener("offline",o,!1),()=>{window.removeEventListener("online",a),window.removeEventListener("offline",o)}}})}onSubscribe(){A(this,et)||this.setEventListener(A(this,Et))}onUnsubscribe(){var t;this.hasListeners()||((t=A(this,et))==null||t.call(this),F(this,et,void 0))}setEventListener(t){var a;F(this,Et,t),(a=A(this,et))==null||a.call(this),F(this,et,t(this.setOnline.bind(this)))}setOnline(t){A(this,Tt)!==t&&(F(this,Tt,t),this.listeners.forEach(o=>{o(t)}))}isOnline(){return A(this,Tt)}},Tt=new WeakMap,et=new WeakMap,Et=new WeakMap,vo),vn=new Kl;function $l(){let e,t;const a=new Promise((r,s)=>{e=r,t=s});a.status="pending",a.catch(()=>{});function o(r){Object.assign(a,r),delete a.resolve,delete a.reject}return a.resolve=r=>{o({status:"fulfilled",value:r}),e(r)},a.reject=r=>{o({status:"rejected",reason:r}),t(r)},a}function Yl(e){return Math.min(1e3*2**e,3e4)}function Ar(e){return(e??"online")==="online"?vn.isOnline():!0}var xr=class extends Error{constructor(e){super("CancelledError"),this.revert=e==null?void 0:e.revert,this.silent=e==null?void 0:e.silent}};function Fn(e){return e instanceof xr}function wr(e){let t=!1,a=0,o=!1,r;const s=$l(),d=y=>{var b;o||(g(new xr(y)),(b=e.abort)==null||b.call(e))},l=()=>{t=!0},m=()=>{t=!1},u=()=>Ir.isFocused()&&(e.networkMode==="always"||vn.isOnline())&&e.canRun(),v=()=>Ar(e.networkMode)&&e.canRun(),h=y=>{var b;o||(o=!0,(b=e.onSuccess)==null||b.call(e,y),r==null||r(),s.resolve(y))},g=y=>{var b;o||(o=!0,(b=e.onError)==null||b.call(e,y),r==null||r(),s.reject(y))},c=()=>new Promise(y=>{var b;r=x=>{(o||u())&&y(x)},(b=e.onPause)==null||b.call(e)}).then(()=>{var y;r=void 0,o||(y=e.onContinue)==null||y.call(e)}),w=()=>{if(o)return;let y;const b=a===0?e.initialPromise:void 0;try{y=b??e.fn()}catch(x){y=Promise.reject(x)}Promise.resolve(y).then(h).catch(x=>{var E;if(o)return;const I=e.retry??(Sn?0:3),k=e.retryDelay??Yl,P=typeof k=="function"?k(a,x):k,T=I===!0||typeof I=="number"&&a<I||typeof I=="function"&&I(a,x);if(t||!T){g(x);return}a++,(E=e.onFail)==null||E.call(e,a,x),Ul(P).then(()=>u()?void 0:c()).then(()=>{t?g(x):w()})})};return{promise:s,cancel:d,continue:()=>(r==null||r(),s),cancelRetry:l,continueRetry:m,canStart:v,start:()=>(v()?w():c().then(w),s)}}var Zl=e=>setTimeout(e,0);function Jl(){let e=[],t=0,a=l=>{l()},o=l=>{l()},r=Zl;const s=l=>{t?e.push(l):r(()=>{a(l)})},d=()=>{const l=e;e=[],l.length&&r(()=>{o(()=>{l.forEach(m=>{a(m)})})})};return{batch:l=>{let m;t++;try{m=l()}finally{t--,t||d()}return m},batchCalls:l=>(...m)=>{s(()=>{l(...m)})},schedule:s,setNotifyFunction:l=>{a=l},setBatchNotifyFunction:l=>{o=l},setScheduler:l=>{r=l}}}var me=Jl(),ut,Io,Sr=(Io=class{constructor(){U(this,ut)}destroy(){this.clearGcTimeout()}scheduleGc(){this.clearGcTimeout(),jl(this.gcTime)&&F(this,ut,setTimeout(()=>{this.optionalRemove()},this.gcTime))}updateGcTime(e){this.gcTime=Math.max(this.gcTime||0,e??(Sn?1/0:5*60*1e3))}clearGcTimeout(){A(this,ut)&&(clearTimeout(A(this,ut)),F(this,ut,void 0))}},ut=new WeakMap,Io),Rt,mt,Ae,pt,re,Yt,ft,Pe,Be,Ao,Xl=(Ao=class extends Sr{constructor(t){super();U(this,Pe);U(this,Rt);U(this,mt);U(this,Ae);U(this,pt);U(this,re);U(this,Yt);U(this,ft);F(this,ft,!1),F(this,Yt,t.defaultOptions),this.setOptions(t.options),this.observers=[],F(this,pt,t.client),F(this,Ae,A(this,pt).getQueryCache()),this.queryKey=t.queryKey,this.queryHash=t.queryHash,F(this,Rt,tc(this.options)),this.state=t.state??A(this,Rt),this.scheduleGc()}get meta(){return this.options.meta}get promise(){var t;return(t=A(this,re))==null?void 0:t.promise}setOptions(t){this.options={...A(this,Yt),...t},this.updateGcTime(this.options.gcTime)}optionalRemove(){!this.observers.length&&this.state.fetchStatus==="idle"&&A(this,Ae).remove(this)}setData(t,a){const o=Ql(this.state.data,t,this.options);return ae(this,Pe,Be).call(this,{data:o,type:"success",dataUpdatedAt:a==null?void 0:a.updatedAt,manual:a==null?void 0:a.manual}),o}setState(t,a){ae(this,Pe,Be).call(this,{type:"setState",state:t,setStateOptions:a})}cancel(t){var o,r;const a=(o=A(this,re))==null?void 0:o.promise;return(r=A(this,re))==null||r.cancel(t),a?a.then(Ce).catch(Ce):Promise.resolve()}destroy(){super.destroy(),this.cancel({silent:!0})}reset(){this.destroy(),this.setState(A(this,Rt))}isActive(){return this.observers.some(t=>Bl(t.options.enabled,this)!==!1)}isDisabled(){return this.getObserversCount()>0?!this.isActive():this.options.queryFn===ba||this.state.dataUpdateCount+this.state.errorUpdateCount===0}isStatic(){return this.getObserversCount()>0?this.observers.some(t=>Xn(t.options.staleTime,this)==="static"):!1}isStale(){return this.getObserversCount()>0?this.observers.some(t=>t.getCurrentResult().isStale):this.state.data===void 0||this.state.isInvalidated}isStaleByTime(t=0){return this.state.data===void 0?!0:t==="static"?!1:this.state.isInvalidated?!0:!Gl(this.state.dataUpdatedAt,t)}onFocus(){var a;const t=this.observers.find(o=>o.shouldFetchOnWindowFocus());t==null||t.refetch({cancelRefetch:!1}),(a=A(this,re))==null||a.continue()}onOnline(){var a;const t=this.observers.find(o=>o.shouldFetchOnReconnect());t==null||t.refetch({cancelRefetch:!1}),(a=A(this,re))==null||a.continue()}addObserver(t){this.observers.includes(t)||(this.observers.push(t),this.clearGcTimeout(),A(this,Ae).notify({type:"observerAdded",query:this,observer:t}))}removeObserver(t){this.observers.includes(t)&&(this.observers=this.observers.filter(a=>a!==t),this.observers.length||(A(this,re)&&(A(this,ft)?A(this,re).cancel({revert:!0}):A(this,re).cancelRetry()),this.scheduleGc()),A(this,Ae).notify({type:"observerRemoved",query:this,observer:t}))}getObserversCount(){return this.observers.length}invalidate(){this.state.isInvalidated||ae(this,Pe,Be).call(this,{type:"invalidate"})}fetch(t,a){var u,v,h;if(this.state.fetchStatus!=="idle"){if(this.state.data!==void 0&&(a!=null&&a.cancelRefetch))this.cancel({silent:!0});else if(A(this,re))return A(this,re).continueRetry(),A(this,re).promise}if(t&&this.setOptions(t),!this.options.queryFn){const g=this.observers.find(c=>c.options.queryFn);g&&this.setOptions(g.options)}const o=new AbortController,r=g=>{Object.defineProperty(g,"signal",{enumerable:!0,get:()=>(F(this,ft,!0),o.signal)})},s=()=>{const g=vr(this.options,a),w=(()=>{const y={client:A(this,pt),queryKey:this.queryKey,meta:this.meta};return r(y),y})();return F(this,ft,!1),this.options.persister?this.options.persister(g,w,this):g(w)},l=(()=>{const g={fetchOptions:a,options:this.options,queryKey:this.queryKey,client:A(this,pt),state:this.state,fetchFn:s};return r(g),g})();(u=this.options.behavior)==null||u.onFetch(l,this),F(this,mt,this.state),(this.state.fetchStatus==="idle"||this.state.fetchMeta!==((v=l.fetchOptions)==null?void 0:v.meta))&&ae(this,Pe,Be).call(this,{type:"fetch",meta:(h=l.fetchOptions)==null?void 0:h.meta});const m=g=>{var c,w,y,b;Fn(g)&&g.silent||ae(this,Pe,Be).call(this,{type:"error",error:g}),Fn(g)||((w=(c=A(this,Ae).config).onError)==null||w.call(c,g,this),(b=(y=A(this,Ae).config).onSettled)==null||b.call(y,this.state.data,g,this)),this.scheduleGc()};return F(this,re,wr({initialPromise:a==null?void 0:a.initialPromise,fn:l.fetchFn,abort:o.abort.bind(o),onSuccess:g=>{var c,w,y,b;if(g===void 0){m(new Error(`${this.queryHash} data is undefined`));return}try{this.setData(g)}catch(x){m(x);return}(w=(c=A(this,Ae).config).onSuccess)==null||w.call(c,g,this),(b=(y=A(this,Ae).config).onSettled)==null||b.call(y,g,this.state.error,this),this.scheduleGc()},onError:m,onFail:(g,c)=>{ae(this,Pe,Be).call(this,{type:"failed",failureCount:g,error:c})},onPause:()=>{ae(this,Pe,Be).call(this,{type:"pause"})},onContinue:()=>{ae(this,Pe,Be).call(this,{type:"continue"})},retry:l.options.retry,retryDelay:l.options.retryDelay,networkMode:l.options.networkMode,canRun:()=>!0})),A(this,re).start()}},Rt=new WeakMap,mt=new WeakMap,Ae=new WeakMap,pt=new WeakMap,re=new WeakMap,Yt=new WeakMap,ft=new WeakMap,Pe=new WeakSet,Be=function(t){const a=o=>{switch(t.type){case"failed":return{...o,fetchFailureCount:t.failureCount,fetchFailureReason:t.error};case"pause":return{...o,fetchStatus:"paused"};case"continue":return{...o,fetchStatus:"fetching"};case"fetch":return{...o,...ec(o.data,this.options),fetchMeta:t.meta??null};case"success":return F(this,mt,void 0),{...o,data:t.data,dataUpdateCount:o.dataUpdateCount+1,dataUpdatedAt:t.dataUpdatedAt??Date.now(),error:null,isInvalidated:!1,status:"success",...!t.manual&&{fetchStatus:"idle",fetchFailureCount:0,fetchFailureReason:null}};case"error":const r=t.error;return Fn(r)&&r.revert&&A(this,mt)?{...A(this,mt),fetchStatus:"idle"}:{...o,error:r,errorUpdateCount:o.errorUpdateCount+1,errorUpdatedAt:Date.now(),fetchFailureCount:o.fetchFailureCount+1,fetchFailureReason:r,fetchStatus:"idle",status:"error"};case"invalidate":return{...o,isInvalidated:!0};case"setState":return{...o,...t.state}}};this.state=a(this.state),me.batch(()=>{this.observers.forEach(o=>{o.onQueryUpdate()}),A(this,Ae).notify({query:this,type:"updated",action:t})})},Ao);function ec(e,t){return{fetchFailureCount:0,fetchFailureReason:null,fetchStatus:Ar(t.networkMode)?"fetching":"paused",...e===void 0&&{error:null,status:"pending"}}}function tc(e){const t=typeof e.initialData=="function"?e.initialData():e.initialData,a=t!==void 0,o=a?typeof e.initialDataUpdatedAt=="function"?e.initialDataUpdatedAt():e.initialDataUpdatedAt:0;return{data:t,dataUpdateCount:0,dataUpdatedAt:a?o??Date.now():0,error:null,errorUpdateCount:0,errorUpdatedAt:0,fetchFailureCount:0,fetchFailureReason:null,fetchMeta:null,isInvalidated:!1,status:a?"success":"pending",fetchStatus:"idle"}}var Fe,xo,nc=(xo=class extends wn{constructor(t={}){super();U(this,Fe);this.config=t,F(this,Fe,new Map)}build(t,a,o){const r=a.queryKey,s=a.queryHash??ga(r,a);let d=this.get(s);return d||(d=new Xl({client:t,queryKey:r,queryHash:s,options:t.defaultQueryOptions(a),state:o,defaultOptions:t.getQueryDefaults(r)}),this.add(d)),d}add(t){A(this,Fe).has(t.queryHash)||(A(this,Fe).set(t.queryHash,t),this.notify({type:"added",query:t}))}remove(t){const a=A(this,Fe).get(t.queryHash);a&&(t.destroy(),a===t&&A(this,Fe).delete(t.queryHash),this.notify({type:"removed",query:t}))}clear(){me.batch(()=>{this.getAll().forEach(t=>{this.remove(t)})})}get(t){return A(this,Fe).get(t)}getAll(){return[...A(this,Fe).values()]}find(t){const a={exact:!0,...t};return this.getAll().find(o=>Ba(a,o))}findAll(t={}){const a=this.getAll();return Object.keys(t).length>0?a.filter(o=>Ba(t,o)):a}notify(t){me.batch(()=>{this.listeners.forEach(a=>{a(t)})})}onFocus(){me.batch(()=>{this.getAll().forEach(t=>{t.onFocus()})})}onOnline(){me.batch(()=>{this.getAll().forEach(t=>{t.onOnline()})})}},Fe=new WeakMap,xo),Ne,ue,ht,_e,Je,wo,ac=(wo=class extends Sr{constructor(t){super();U(this,_e);U(this,Ne);U(this,ue);U(this,ht);this.mutationId=t.mutationId,F(this,ue,t.mutationCache),F(this,Ne,[]),this.state=t.state||oc(),this.setOptions(t.options),this.scheduleGc()}setOptions(t){this.options=t,this.updateGcTime(this.options.gcTime)}get meta(){return this.options.meta}addObserver(t){A(this,Ne).includes(t)||(A(this,Ne).push(t),this.clearGcTimeout(),A(this,ue).notify({type:"observerAdded",mutation:this,observer:t}))}removeObserver(t){F(this,Ne,A(this,Ne).filter(a=>a!==t)),this.scheduleGc(),A(this,ue).notify({type:"observerRemoved",mutation:this,observer:t})}optionalRemove(){A(this,Ne).length||(this.state.status==="pending"?this.scheduleGc():A(this,ue).remove(this))}continue(){var t;return((t=A(this,ht))==null?void 0:t.continue())??this.execute(this.state.variables)}async execute(t){var s,d,l,m,u,v,h,g,c,w,y,b,x,I,k,P,T,E,j,D;const a=()=>{ae(this,_e,Je).call(this,{type:"continue"})};F(this,ht,wr({fn:()=>this.options.mutationFn?this.options.mutationFn(t):Promise.reject(new Error("No mutationFn found")),onFail:(q,B)=>{ae(this,_e,Je).call(this,{type:"failed",failureCount:q,error:B})},onPause:()=>{ae(this,_e,Je).call(this,{type:"pause"})},onContinue:a,retry:this.options.retry??0,retryDelay:this.options.retryDelay,networkMode:this.options.networkMode,canRun:()=>A(this,ue).canRun(this)}));const o=this.state.status==="pending",r=!A(this,ht).canStart();try{if(o)a();else{ae(this,_e,Je).call(this,{type:"pending",variables:t,isPaused:r}),await((d=(s=A(this,ue).config).onMutate)==null?void 0:d.call(s,t,this));const B=await((m=(l=this.options).onMutate)==null?void 0:m.call(l,t));B!==this.state.context&&ae(this,_e,Je).call(this,{type:"pending",context:B,variables:t,isPaused:r})}const q=await A(this,ht).start();return await((v=(u=A(this,ue).config).onSuccess)==null?void 0:v.call(u,q,t,this.state.context,this)),await((g=(h=this.options).onSuccess)==null?void 0:g.call(h,q,t,this.state.context)),await((w=(c=A(this,ue).config).onSettled)==null?void 0:w.call(c,q,null,this.state.variables,this.state.context,this)),await((b=(y=this.options).onSettled)==null?void 0:b.call(y,q,null,t,this.state.context)),ae(this,_e,Je).call(this,{type:"success",data:q}),q}catch(q){try{throw await((I=(x=A(this,ue).config).onError)==null?void 0:I.call(x,q,t,this.state.context,this)),await((P=(k=this.options).onError)==null?void 0:P.call(k,q,t,this.state.context)),await((E=(T=A(this,ue).config).onSettled)==null?void 0:E.call(T,void 0,q,this.state.variables,this.state.context,this)),await((D=(j=this.options).onSettled)==null?void 0:D.call(j,void 0,q,t,this.state.context)),q}finally{ae(this,_e,Je).call(this,{type:"error",error:q})}}finally{A(this,ue).runNext(this)}}},Ne=new WeakMap,ue=new WeakMap,ht=new WeakMap,_e=new WeakSet,Je=function(t){const a=o=>{switch(t.type){case"failed":return{...o,failureCount:t.failureCount,failureReason:t.error};case"pause":return{...o,isPaused:!0};case"continue":return{...o,isPaused:!1};case"pending":return{...o,context:t.context,data:void 0,failureCount:0,failureReason:null,error:null,isPaused:t.isPaused,status:"pending",variables:t.variables,submittedAt:Date.now()};case"success":return{...o,data:t.data,failureCount:0,failureReason:null,error:null,status:"success",isPaused:!1};case"error":return{...o,data:void 0,error:t.error,failureCount:o.failureCount+1,failureReason:t.error,isPaused:!1,status:"error"}}};this.state=a(this.state),me.batch(()=>{A(this,Ne).forEach(o=>{o.onMutationUpdate(t)}),A(this,ue).notify({mutation:this,type:"updated",action:t})})},wo);function oc(){return{context:void 0,data:void 0,error:null,failureCount:0,failureReason:null,isPaused:!1,status:"idle",variables:void 0,submittedAt:0}}var Ue,Te,Zt,So,rc=(So=class extends wn{constructor(t={}){super();U(this,Ue);U(this,Te);U(this,Zt);this.config=t,F(this,Ue,new Set),F(this,Te,new Map),F(this,Zt,0)}build(t,a,o){const r=new ac({mutationCache:this,mutationId:++nn(this,Zt)._,options:t.defaultMutationOptions(a),state:o});return this.add(r),r}add(t){A(this,Ue).add(t);const a=rn(t);
98if(typeof a=="string"){const o=A(this,Te).get(a);o?o.push(t):A(this,Te).set(a,[t])}this.notify({type:"added",mutation:t})}remove(t){if(A(this,Ue).delete(t)){const a=rn(t);if(typeof a=="string"){const o=A(this,Te).get(a);if(o)if(o.length>1){const r=o.indexOf(t);r!==-1&&o.splice(r,1)}else o[0]===t&&A(this,Te).delete(a)}}this.notify({type:"removed",mutation:t})}canRun(t){const a=rn(t);if(typeof a=="string"){const o=A(this,Te).get(a),r=o==null?void 0:o.find(s=>s.state.status==="pending");return!r||r===t}else return!0}runNext(t){var o;const a=rn(t);if(typeof a=="string"){const r=(o=A(this,Te).get(a))==null?void 0:o.find(s=>s!==t&&s.state.isPaused);return(r==null?void 0:r.continue())??Promise.resolve()}else return Promise.resolve()}clear(){me.batch(()=>{A(this,Ue).forEach(t=>{this.notify({type:"removed",mutation:t})}),A(this,Ue).clear(),A(this,Te).clear()})}getAll(){return Array.from(A(this,Ue))}find(t){const a={exact:!0,...t};return this.getAll().find(o=>Ua(a,o))}findAll(t={}){return this.getAll().filter(a=>Ua(t,a))}notify(t){me.batch(()=>{this.listeners.forEach(a=>{a(t)})})}resumePausedMutations(){const t=this.getAll().filter(a=>a.state.isPaused);return me.batch(()=>Promise.all(t.map(a=>a.continue().catch(Ce))))}},Ue=new WeakMap,Te=new WeakMap,Zt=new WeakMap,So);function rn(e){var t;return(t=e.options.scope)==null?void 0:t.id}function Ha(e){return{onFetch:(t,a)=>{var v,h,g,c,w;const o=t.options,r=(g=(h=(v=t.fetchOptions)==null?void 0:v.meta)==null?void 0:h.fetchMore)==null?void 0:g.direction,s=((c=t.state.data)==null?void 0:c.pages)||[],d=((w=t.state.data)==null?void 0:w.pageParams)||[];let l={pages:[],pageParams:[]},m=0;const u=async()=>{let y=!1;const b=k=>{Object.defineProperty(k,"signal",{enumerable:!0,get:()=>(t.signal.aborted?y=!0:t.signal.addEventListener("abort",()=>{y=!0}),t.signal)})},x=vr(t.options,t.fetchOptions),I=async(k,P,T)=>{if(y)return Promise.reject();if(P==null&&k.pages.length)return Promise.resolve(k);const j=(()=>{const G={client:t.client,queryKey:t.queryKey,pageParam:P,direction:T?"backward":"forward",meta:t.options.meta};return b(G),G})(),D=await x(j),{maxPages:q}=t.options,B=T?Hl:Vl;return{pages:B(k.pages,D,q),pageParams:B(k.pageParams,P,q)}};if(r&&s.length){const k=r==="backward",P=k?ic:Wa,T={pages:s,pageParams:d},E=P(o,T);l=await I(T,E,k)}else{const k=e??s.length;do{const P=m===0?d[0]??o.initialPageParam:Wa(o,l);if(m>0&&P==null)break;l=await I(l,P),m++}while(m<k)}return l};t.options.persister?t.fetchFn=()=>{var y,b;return(b=(y=t.options).persister)==null?void 0:b.call(y,u,{client:t.client,queryKey:t.queryKey,meta:t.options.meta,signal:t.signal},a)}:t.fetchFn=u}}}function Wa(e,{pages:t,pageParams:a}){const o=t.length-1;return t.length>0?e.getNextPageParam(t[o],t,a[o],a):void 0}function ic(e,{pages:t,pageParams:a}){var o;return t.length>0?(o=e.getPreviousPageParam)==null?void 0:o.call(e,t[0],t,a[0],a):void 0}var X,tt,nt,Ot,Mt,at,qt,Dt,ko,sc=(ko=class{constructor(e={}){U(this,X);U(this,tt);U(this,nt);U(this,Ot);U(this,Mt);U(this,at);U(this,qt);U(this,Dt);F(this,X,e.queryCache||new nc),F(this,tt,e.mutationCache||new rc),F(this,nt,e.defaultOptions||{}),F(this,Ot,new Map),F(this,Mt,new Map),F(this,at,0)}mount(){nn(this,at)._++,A(this,at)===1&&(F(this,qt,Ir.subscribe(async e=>{e&&(await this.resumePausedMutations(),A(this,X).onFocus())})),F(this,Dt,vn.subscribe(async e=>{e&&(await this.resumePausedMutations(),A(this,X).onOnline())})))}unmount(){var e,t;nn(this,at)._--,A(this,at)===0&&((e=A(this,qt))==null||e.call(this),F(this,qt,void 0),(t=A(this,Dt))==null||t.call(this),F(this,Dt,void 0))}isFetching(e){return A(this,X).findAll({...e,fetchStatus:"fetching"}).length}isMutating(e){return A(this,tt).findAll({...e,status:"pending"}).length}getQueryData(e){var a;const t=this.defaultQueryOptions({queryKey:e});return(a=A(this,X).get(t.queryHash))==null?void 0:a.state.data}ensureQueryData(e){const t=this.defaultQueryOptions(e),a=A(this,X).build(this,t),o=a.state.data;return o===void 0?this.fetchQuery(e):(e.revalidateIfStale&&a.isStaleByTime(Xn(t.staleTime,a))&&this.prefetchQuery(t),Promise.resolve(o))}getQueriesData(e){return A(this,X).findAll(e).map(({queryKey:t,state:a})=>{const o=a.data;return[t,o]})}setQueryData(e,t,a){const o=this.defaultQueryOptions({queryKey:e}),r=A(this,X).get(o.queryHash),s=r==null?void 0:r.state.data,d=zl(t,s);if(d!==void 0)return A(this,X).build(this,o).setData(d,{...a,manual:!0})}
98setQueriesData(e,t,a){return me.batch(()=>A(this,X).findAll(e).map(({queryKey:o})=>[o,this.setQueryData(o,t,a)]))}getQueryState(e){var a;const t=this.defaultQueryOptions({queryKey:e});return(a=A(this,X).get(t.queryHash))==null?void 0:a.state}removeQueries(e){const t=A(this,X);me.batch(()=>{t.findAll(e).forEach(a=>{t.remove(a)})})}resetQueries(e,t){const a=A(this,X);return me.batch(()=>(a.findAll(e).forEach(o=>{o.reset()}),this.refetchQueries({type:"active",...e},t)))}cancelQueries(e,t={}){const a={revert:!0,...t},o=me.batch(()=>A(this,X).findAll(e).map(r=>r.cancel(a)));return Promise.all(o).then(Ce).catch(Ce)}invalidateQueries(e,t={}){return me.batch(()=>(A(this,X).findAll(e).forEach(a=>{a.invalidate()}),(e==null?void 0:e.refetchType)==="none"?Promise.resolve():this.refetchQueries({...e,type:(e==null?void 0:e.refetchType)??(e==null?void 0:e.type)??"active"},t)))}refetchQueries(e,t={}){const a={...t,cancelRefetch:t.cancelRefetch??!0},o=me.batch(()=>A(this,X).findAll(e).filter(r=>!r.isDisabled()&&!r.isStatic()).map(r=>{let s=r.fetch(void 0,a);return a.throwOnError||(s=s.catch(Ce)),r.state.fetchStatus==="paused"?Promise.resolve():s}));return Promise.all(o).then(Ce)}fetchQuery(e){const t=this.defaultQueryOptions(e);t.retry===void 0&&(t.retry=!1);const a=A(this,X).build(this,t);return a.isStaleByTime(Xn(t.staleTime,a))?a.fetch(t):Promise.resolve(a.state.data)}prefetchQuery(e){return this.fetchQuery(e).then(Ce).catch(Ce)}fetchInfiniteQuery(e){return e.behavior=Ha(e.pages),this.fetchQuery(e)}prefetchInfiniteQuery(e){return this.fetchInfiniteQuery(e).then(Ce).catch(Ce)}ensureInfiniteQueryData(e){return e.behavior=Ha(e.pages),this.ensureQueryData(e)}resumePausedMutations(){return vn.isOnline()?A(this,tt).resumePausedMutations():Promise.resolve()}getQueryCache(){return A(this,X)}getMutationCache(){return A(this,tt)}getDefaultOptions(){return A(this,nt)}setDefaultOptions(e){F(this,nt,e)}setQueryDefaults(e,t){A(this,Ot).set(Kt(e),{queryKey:e,defaultOptions:t})}getQueryDefaults(e){const t=[...A(this,Ot).values()],a={};return t.forEach(o=>{$t(e,o.queryKey)&&Object.assign(a,o.defaultOptions)}),a}setMutationDefaults(e,t){A(this,Mt).set(Kt(e),{mutationKey:e,defaultOptions:t})}getMutationDefaults(e){const t=[...A(this,Mt).values()],a={};return t.forEach(o=>{$t(e,o.mutationKey)&&Object.assign(a,o.defaultOptions)}),a}defaultQueryOptions(e){if(e._defaulted)return e;const t={...A(this,nt).queries,...this.getQueryDefaults(e.queryKey),...e,_defaulted:!0};return t.queryHash||(t.queryHash=ga(t.queryKey,t)),t.refetchOnReconnect===void 0&&(t.refetchOnReconnect=t.networkMode!=="always"),t.throwOnError===void 0&&(t.throwOnError=!!t.suspense),!t.networkMode&&t.persister&&(t.networkMode="offlineFirst"),t.queryFn===ba&&(t.enabled=!1),t}defaultMutationOptions(e){return e!=null&&e._defaulted?e:{...A(this,nt).mutations,...(e==null?void 0:e.mutationKey)&&this.getMutationDefaults(e.mutationKey),...e,_defaulted:!0}}clear(){A(this,X).clear(),A(this,tt).clear()}},X=new WeakMap,tt=new WeakMap,nt=new WeakMap,Ot=new WeakMap,Mt=new WeakMap,at=new WeakMap,qt=new WeakMap,Dt=new WeakMap,ko),kr=i.createContext(void 0),Mg=e=>{const t=i.useContext(kr);if(!t)throw new Error("No QueryClient set, use QueryClientProvider to set one");return t},lc=({client:e,children:t})=>(i.useEffect(()=>(e.mount(),()=>{e.unmount()}),[e]),n.jsx(kr.Provider,{value:e,children:t})),cc=typeof Element<"u",dc=typeof Map=="function",uc=typeof Set=="function",mc=typeof ArrayBuffer=="function"&&!!ArrayBuffer.isView;function fn(e,t){if(e===t)return!0;if(e&&t&&typeof e=="object"&&typeof t=="object"){if(e.constructor!==t.constructor)return!1;var a,o,r;if(Array.isArray(e)){if(a=e.length,a!=t.length)return!1;for(o=a;o--!==0;)if(!fn(e[o],t[o]))return!1;return!0}var s;if(dc&&e instanceof Map&&t instanceof Map){if(e.size!==t.size)return!1;for(s=e.entries();!(o=s.next()).done;)if(!t.has(o.value[0]))return!1;for(s=e.entries();!(o=s.next()).done;)if(!fn(o.value[1],t.get(o.value[0])))return!1;return!0}if(uc&&e instanceof Set&&t instanceof Set){if(e.size!==t.size)return!1;for(s=e.entries();!(o=s.next()).done;)if(!t.has(o.value[0]))return!1;return!0}if(mc&&ArrayBuffer.isView(e)&&ArrayBuffer.isView(t)){if(a=e.length,a!=t.length)return!1;for(o=a;o--!==0;)if(e[o]!==t[o])return!1;return!0}if(e.constructor===RegExp)return e.source===t.source&&e.flags===t.flags;if(e.valueOf!==Object.prototype.valueOf&&typeof e.valueOf=="function"&&typeof t.valueOf=="function")return e.valueOf()===t.valueOf();if(e.toString!==Object.prototype.toString&&typeof e.toString=="function"&&typeof t.toString=="function")return e.toString()===t.toString();if(r=Object.keys(e),a=r.length,a!==Object.keys(t).length)return!1;for(o=a;o--!==0;)if(!Object.prototype.hasOwnProperty.call(t,r[o]))return!1;if(cc&&e instanceof Element)return!1;for(o=a;o--!==0;)if(!((r[o]==="_owner"||r[o]==="__v"||r[o]==="__o")&&e.$$typeof)&&!fn(e[r[o]],t[r[o]]))return!1;return!0}return e!==e&&t!==t}var pc=function(t,a){try{return fn(t,a)}catch(o){if((o.message||"").match(/stack|recursion/i))return console.warn("react-fast-compare cannot handle circular refs"),!1;throw o}};const fc=la(pc);var hc=function(e,t,a,o,r,s,d,l){if(!e){var m;if(t===void 0)m=new Error("Minified exception occurred; use the non-minified dev environment for the full error message and additional helpful warnings.");else{var u=[a,o,r,s,d,l],v=0;m=new Error(t.replace(/%s/g,function(){return u[v++]})),m.name="Invariant Violation"}throw m.framesToPop=1,m}},gc=hc;const Ka=la(gc);var bc=function(t,a,o,r){var s=o?o.call(r,t,a):void 0;if(s!==void 0)return!!s;
98if(t===a)return!0;if(typeof t!="object"||!t||typeof a!="object"||!a)return!1;var d=Object.keys(t),l=Object.keys(a);if(d.length!==l.length)return!1;for(var m=Object.prototype.hasOwnProperty.bind(a),u=0;u<d.length;u++){var v=d[u];if(!m(v))return!1;var h=t[v],g=a[v];if(s=o?o.call(r,h,g,v):void 0,s===!1||s===void 0&&h!==g)return!1}return!0};const yc=la(bc);var Cr=(e=>(e.BASE="base",e.BODY="body",e.HEAD="head",e.HTML="html",e.LINK="link",e.META="meta",e.NOSCRIPT="noscript",e.SCRIPT="script",e.STYLE="style",e.TITLE="title",e.FRAGMENT="Symbol(react.fragment)",e))(Cr||{}),Nn={link:{rel:["amphtml","canonical","alternate"]},script:{type:["application/ld+json"]},meta:{charset:"",name:["generator","robots","description"],property:["og:type","og:title","og:url","og:image","og:image:alt","og:description","twitter:url","twitter:title","twitter:description","twitter:image","twitter:image:alt","twitter:card","twitter:site"]}},$a=Object.values(Cr),ya={accesskey:"accessKey",charset:"charSet",class:"className",contenteditable:"contentEditable",contextmenu:"contextMenu","http-equiv":"httpEquiv",itemprop:"itemProp",tabindex:"tabIndex"},vc=Object.entries(ya).reduce((e,[t,a])=>(e[a]=t,e),{}),Oe="data-rh",kt={DEFAULT_TITLE:"defaultTitle",DEFER:"defer",ENCODE_SPECIAL_CHARACTERS:"encodeSpecialCharacters",ON_CHANGE_CLIENT_STATE:"onChangeClientState",TITLE_TEMPLATE:"titleTemplate",PRIORITIZE_SEO_TAGS:"prioritizeSeoTags"},Ct=(e,t)=>{for(let a=e.length-1;a>=0;a-=1){const o=e[a];if(Object.prototype.hasOwnProperty.call(o,t))return o[t]}return null},Ic=e=>{let t=Ct(e,"title");const a=Ct(e,kt.TITLE_TEMPLATE);if(Array.isArray(t)&&(t=t.join("")),a&&t)return a.replace(/%s/g,()=>t);const o=Ct(e,kt.DEFAULT_TITLE);return t||o||void 0},Ac=e=>Ct(e,kt.ON_CHANGE_CLIENT_STATE)||(()=>{}),_n=(e,t)=>t.filter(a=>typeof a[e]<"u").map(a=>a[e]).reduce((a,o)=>({...a,...o}),{}),xc=(e,t)=>t.filter(a=>typeof a.base<"u").map(a=>a.base).reverse().reduce((a,o)=>{if(!a.length){const r=Object.keys(o);for(let s=0;s<r.length;s+=1){const l=r[s].toLowerCase();if(e.indexOf(l)!==-1&&o[l])return a.concat(o)}}return a},[]),wc=e=>console&&typeof console.warn=="function"&&console.warn(e),Ut=(e,t,a)=>{const o={};return a.filter(r=>Array.isArray(r[e])?!0:(typeof r[e]<"u"&&wc(`Helmet: ${e} should be of type "Array". Instead found type "${typeof r[e]}"`),!1)).map(r=>r[e]).reverse().reduce((r,s)=>{const d={};s.filter(m=>{let u;const v=Object.keys(m);for(let g=0;g<v.length;g+=1){const c=v[g],w=c.toLowerCase();t.indexOf(w)!==-1&&!(u==="rel"&&m[u].toLowerCase()==="canonical")&&!(w==="rel"&&m[w].toLowerCase()==="stylesheet")&&(u=w),t.indexOf(c)!==-1&&(c==="innerHTML"||c==="cssText"||c==="itemprop")&&(u=c)}if(!u||!m[u])return!1;const h=m[u].toLowerCase();return o[u]||(o[u]={}),d[u]||(d[u]={}),o[u][h]?!1:(d[u][h]=!0,!0)}).reverse().forEach(m=>r.push(m));const l=Object.keys(d);for(let m=0;m<l.length;m+=1){const u=l[m],v={...o[u],...d[u]};o[u]=v}return r},[]).reverse()},Sc=(e,t)=>{if(Array.isArray(e)&&e.length){for(let a=0;a<e.length;a+=1)if(e[a][t])return!0}return!1},kc=e=>({baseTag:xc(["href"],e),bodyAttributes:_n("bodyAttributes",e),defer:Ct(e,kt.DEFER),encode:Ct(e,kt.ENCODE_SPECIAL_CHARACTERS),htmlAttributes:_n("htmlAttributes",e),linkTags:Ut("link",["rel","href"],e),metaTags:Ut("meta",["name","charset","http-equiv","property","itemprop"],e),noscriptTags:Ut("noscript",["innerHTML"],e),onChangeClientState:Ac(e),scriptTags:Ut("script",["src","innerHTML"],e),styleTags:Ut("style",["cssText"],e),title:Ic(e),titleAttributes:_n("titleAttributes",e),prioritizeSeoTags:Sc(e,kt.PRIORITIZE_SEO_TAGS)}),Pr=e=>Array.isArray(e)?e.join(""):e,Cc=(e,t)=>{const a=Object.keys(e);for(let o=0;o<a.length;o+=1)if(t[a[o]]&&t[a[o]].includes(e[a[o]]))return!0;return!1},zn=(e,t)=>Array.isArray(e)?e.reduce((a,o)=>(Cc(o,t)?a.priority.push(o):a.default.push(o),a),{priority:[],default:[]}):{default:e,priority:[]},Ya=(e,t)=>({...e,[t]:void 0}),Pc=["noscript","script","style"],ta=(e,t=!0)=>t===!1?String(e):String(e).replace(/&/g,"&").replace(/</g,"<").replace(/>/g,">").replace(/"/g,""").replace(/'/g,"'"),Tr=e=>Object.keys(e).reduce((t,a)=>{const o=typeof e[a]<"u"?`${a}="${e[a]}"`:`${a}`;return t?`${t} ${o}`:o},""),Tc=(e,t,a,o)=>{const r=Tr(a),s=Pr(t);return r?`<${e} ${Oe}="true" ${r}>${ta(s,o)}</${e}>`:`<${e} ${Oe}="true">${ta(s,o)}</${e}>`},Ec=(e,t,a=!0)=>t.reduce((o,r)=>{const s=r,d=Object.keys(s).filter(u=>!(u==="innerHTML"||u==="cssText")).reduce((u,v)=>{const h=typeof s[v]>"u"?v:`${v}="${ta(s[v],a)}"`;return u?`${u} ${h}`:h},""),l=s.innerHTML||s.cssText||"",m=Pc.indexOf(e)===-1;return`${o}<${e} ${Oe}="true" ${d}${m?"/>":`>${l}</${e}>`}`},""),Er=(e,t={})=>Object.keys(e).reduce((a,o)=>{const r=ya[o];return a[r||o]=e[o],a},t),Rc=(e,t,a)=>{const o={key:t,[Oe]:!0},r=Er(a,o);return[C.createElement("title",r,t)]},hn=(e,t)=>t.map((a,o)=>{const r={key:o,[Oe]:!0};
98return Object.keys(a).forEach(s=>{const l=ya[s]||s;if(l==="innerHTML"||l==="cssText"){const m=a.innerHTML||a.cssText;r.dangerouslySetInnerHTML={__html:m}}else r[l]=a[s]}),C.createElement(e,r)}),Ie=(e,t,a=!0)=>{switch(e){case"title":return{toComponent:()=>Rc(e,t.title,t.titleAttributes),toString:()=>Tc(e,t.title,t.titleAttributes,a)};case"bodyAttributes":case"htmlAttributes":return{toComponent:()=>Er(t),toString:()=>Tr(t)};default:return{toComponent:()=>hn(e,t),toString:()=>Ec(e,t,a)}}},Oc=({metaTags:e,linkTags:t,scriptTags:a,encode:o})=>{const r=zn(e,Nn.meta),s=zn(t,Nn.link),d=zn(a,Nn.script);return{priorityMethods:{toComponent:()=>[...hn("meta",r.priority),...hn("link",s.priority),...hn("script",d.priority)],toString:()=>`${Ie("meta",r.priority,o)} ${Ie("link",s.priority,o)} ${Ie("script",d.priority,o)}`},metaTags:r.default,linkTags:s.default,scriptTags:d.default}},Mc=e=>{const{baseTag:t,bodyAttributes:a,encode:o=!0,htmlAttributes:r,noscriptTags:s,styleTags:d,title:l="",titleAttributes:m,prioritizeSeoTags:u}=e;let{linkTags:v,metaTags:h,scriptTags:g}=e,c={toComponent:()=>{},toString:()=>""};return u&&({priorityMethods:c,linkTags:v,metaTags:h,scriptTags:g}=Oc(e)),{priority:c,base:Ie("base",t,o),bodyAttributes:Ie("bodyAttributes",a,o),htmlAttributes:Ie("htmlAttributes",r,o),link:Ie("link",v,o),meta:Ie("meta",h,o),noscript:Ie("noscript",s,o),script:Ie("script",g,o),style:Ie("style",d,o),title:Ie("title",{title:l,titleAttributes:m},o)}},na=Mc,sn=[],Rr=!!(typeof window<"u"&&window.document&&window.document.createElement),aa=class{constructor(e,t){qe(this,"instances",[]);qe(this,"canUseDOM",Rr);qe(this,"context");qe(this,"value",{setHelmet:e=>{this.context.helmet=e},helmetInstances:{get:()=>this.canUseDOM?sn:this.instances,add:e=>{(this.canUseDOM?sn:this.instances).push(e)},remove:e=>{const t=(this.canUseDOM?sn:this.instances).indexOf(e);(this.canUseDOM?sn:this.instances).splice(t,1)}}});this.context=e,this.canUseDOM=t||!1,t||(e.helmet=na({baseTag:[],bodyAttributes:{},encodeSpecialCharacters:!0,htmlAttributes:{},linkTags:[],metaTags:[],noscriptTags:[],scriptTags:[],styleTags:[],title:"",titleAttributes:{}}))}},qc={},Or=C.createContext(qc),gt,Mr=(gt=class extends i.Component{constructor(a){super(a);qe(this,"helmetData");this.helmetData=new aa(this.props.context||{},gt.canUseDOM)}render(){return C.createElement(Or.Provider,{value:this.helmetData.value},this.props.children)}},qe(gt,"canUseDOM",Rr),gt),At=(e,t)=>{const a=document.head||document.querySelector("head"),o=a.querySelectorAll(`${e}[${Oe}]`),r=[].slice.call(o),s=[];let d;return t&&t.length&&t.forEach(l=>{const m=document.createElement(e);for(const u in l)if(Object.prototype.hasOwnProperty.call(l,u))if(u==="innerHTML")m.innerHTML=l.innerHTML;else if(u==="cssText")m.styleSheet?m.styleSheet.cssText=l.cssText:m.appendChild(document.createTextNode(l.cssText));else{const v=u,h=typeof l[v]>"u"?"":l[v];m.setAttribute(u,h)}m.setAttribute(Oe,"true"),r.some((u,v)=>(d=v,m.isEqualNode(u)))?r.splice(d,1):s.push(m)}),r.forEach(l=>{var m;return(m=l.parentNode)==null?void 0:m.removeChild(l)}),s.forEach(l=>a.appendChild(l)),{oldTags:r,newTags:s}},oa=(e,t)=>{const a=document.getElementsByTagName(e)[0];if(!a)return;const o=a.getAttribute(Oe),r=o?o.split(","):[],s=[...r],d=Object.keys(t);for(const l of d){const m=t[l]||"";a.getAttribute(l)!==m&&a.setAttribute(l,m),r.indexOf(l)===-1&&r.push(l);const u=s.indexOf(l);u!==-1&&s.splice(u,1)}for(let l=s.length-1;l>=0;l-=1)a.removeAttribute(s[l]);r.length===s.length?a.removeAttribute(Oe):a.getAttribute(Oe)!==d.join(",")&&a.setAttribute(Oe,d.join(","))},Dc=(e,t)=>{typeof e<"u"&&document.title!==e&&(document.title=Pr(e)),oa("title",t)},Za=(e,t)=>{const{baseTag:a,bodyAttributes:o,htmlAttributes:r,linkTags:s,metaTags:d,noscriptTags:l,onChangeClientState:m,scriptTags:u,styleTags:v,title:h,titleAttributes:g}=e;oa("body",o),oa("html",r),Dc(h,g);const c={baseTag:At("base",a),linkTags:At("link",s),metaTags:At("meta",d),noscriptTags:At("noscript",l),scriptTags:At("script",u),styleTags:At("style",v)},w={},y={};Object.keys(c).forEach(b=>{const{newTags:x,oldTags:I}=c[b];x.length&&(w[b]=x),I.length&&(y[b]=c[b].oldTags)}),t&&t(),m(e,w,y)},Qt=null,Lc=e=>{Qt&&cancelAnimationFrame(Qt),e.defer?Qt=requestAnimationFrame(()=>{Za(e,()=>{Qt=null})}):(Za(e),Qt=null)},Fc=Lc,Ja=class extends i.Component{constructor(){super(...arguments);qe(this,"rendered",!1)}shouldComponentUpdate(t){return!yc(t,this.props)}componentDidUpdate(){this.emitChange()}componentWillUnmount(){const{helmetInstances:t}=this.props.context;t.remove(this),this.emitChange()}emitChange(){const{helmetInstances:t,setHelmet:a}=this.props.context;let o=null;const r=kc(t.get().map(s=>{const d={...s.props};return delete d.context,d}));Mr.canUseDOM?Fc(r):na&&(o=na(r)),a(o)}init(){if(this.rendered)return;this.rendered=!0;const{helmetInstances:t}=this.props.context;t.add(this),this.emitChange()}render(){return this.init(),null}},Hn,Nc=(Hn=class extends i.Component{shouldComponentUpdate(e){return!fc(Ya(this.props,"helmetData"),Ya(e,"helmetData"))}mapNestedChildrenToProps(e,t){if(!t)return null;switch(e.type){case"script":case"noscript":return{innerHTML:t};case"style":return{cssText:t};default:throw new Error(`<${e.type} /> elements are self-closing and can not contain children. Refer to our API for more information.`)}}flattenArrayTypeChildren(e,t,a,o){return{...t,[e.type]:[...t[e.type]||[],{...a,...this.mapNestedChildrenToProps(e,o)}]}}mapObjectTypeChildren(e,t,a,o){switch(e.type){case"title":return{...t,[e.type]:o,titleAttributes:{...a}};case"body":return{...t,bodyAttributes:{...a}};case"html":return{...t,htmlAttributes:{...a}};default:return{...t,[e.type]:{...a}}}}mapArrayTypeChildrenToProps(e,t){let a={...t};return Object.keys(e).forEach(o=>{a={...a,[o]:e[o]}}),a}
98warnOnInvalidChildren(e,t){return Ka($a.some(a=>e.type===a),typeof e.type=="function"?"You may be attempting to nest <Helmet> components within each other, which is not allowed. Refer to our API for more information.":`Only elements types ${$a.join(", ")} are allowed. Helmet does not support rendering <${e.type}> elements. Refer to our API for more information.`),Ka(!t||typeof t=="string"||Array.isArray(t)&&!t.some(a=>typeof a!="string"),`Helmet expects a string as a child of <${e.type}>. Did you forget to wrap your children in braces? ( <${e.type}>{\`\`}</${e.type}> ) Refer to our API for more information.`),!0}mapChildrenToProps(e,t){let a={};return C.Children.forEach(e,o=>{if(!o||!o.props)return;const{children:r,...s}=o.props,d=Object.keys(s).reduce((m,u)=>(m[vc[u]||u]=s[u],m),{});let{type:l}=o;switch(typeof l=="symbol"?l=l.toString():this.warnOnInvalidChildren(o,r),l){case"Symbol(react.fragment)":t=this.mapChildrenToProps(r,t);break;case"link":case"meta":case"noscript":case"script":case"style":a=this.flattenArrayTypeChildren(o,a,d,r);break;default:t=this.mapObjectTypeChildren(o,t,d,r);break}}),this.mapArrayTypeChildrenToProps(a,t)}render(){const{children:e,...t}=this.props;let a={...t},{helmetData:o}=t;if(e&&(a=this.mapChildrenToProps(e,a)),o&&!(o instanceof aa)){const r=o;o=new aa(r.context,!0),delete a.helmetData}return o?C.createElement(Ja,{...a,context:o.value}):C.createElement(Or.Consumer,null,r=>C.createElement(Ja,{...a,context:r}))}},qe(Hn,"defaultProps",{defer:!0,encodeSpecialCharacters:!0,prioritizeSeoTags:!1}),Hn);const _c=()=>{const{pathname:e,hash:t}=Po();return i.useEffect(()=>{if(t){const a=t.replace("#","");let o=0;const r=60,s=()=>{const l=document.getElementById(a);return l?(l.scrollIntoView({behavior:"smooth",block:"start"}),!0):!1};if(s())return;const d=window.setInterval(()=>{o+=1,(s()||o>=r)&&window.clearInterval(d)},50);return()=>window.clearInterval(d)}window.scrollTo({top:0,behavior:"instant"})},[e,t]),null};class zc extends i.Component{constructor(a){super(a);qe(this,"handleRetry",()=>{this.setState(a=>({hasError:!1,error:void 0,retryCount:a.retryCount+1}))});this.state={hasError:!1,retryCount:0}}static getDerivedStateFromError(a){return{hasError:!0,error:a}}componentDidCatch(a,o){var r,s,d;console.error("React ErrorBoundary caught:",a,o);try{const l=JSON.stringify({type:"react_error_boundary",message:((r=a==null?void 0:a.message)==null?void 0:r.substring(0,500))||"Unknown error",stack:((s=a==null?void 0:a.stack)==null?void 0:s.substring(0,1e3))||"",componentStack:((d=o==null?void 0:o.componentStack)==null?void 0:d.substring(0,500))||"",url:window.location.href,ua:navigator.userAgent,ts:new Date().toISOString()});navigator.sendBeacon("https://ikhnpszzlgazzfhvkfsn.supabase.co/functions/v1/site-health-monitor",l)}catch{}}render(){var a;return this.state.hasError?this.state.retryCount===0?(setTimeout(()=>this.handleRetry(),500),n.jsxs("div",{style:{minHeight:"100vh",display:"flex",alignItems:"center",justifyContent:"center",background:"#0a0a0f"},children:[n.jsx("div",{style:{animation:"spin 1s linear infinite",width:48,height:48},children:n.jsx("svg",{width:"48",height:"48",viewBox:"0 0 48 48",children:n.jsx("circle",{cx:"24",cy:"24",r:"20",stroke:"hsl(var(--primary))",strokeWidth:"3",fill:"none",strokeDasharray:"80 40",strokeLinecap:"round"})})}),n.jsx("style",{children:"@keyframes spin{to{transform:rotate(360deg)}}"})]})):n.jsxs("div",{style:{minHeight:"100vh",display:"flex",alignItems:"center",justifyContent:"center",background:"#0a0a0f",color:"#e2e8f0",fontFamily:"Inter, system-ui, sans-serif",textAlign:"center",padding:"20px",flexDirection:"column",gap:"12px"},children:[n.jsx("p",{style:{fontSize:"24px"},children:"â ï¸ Something went wrong"}),n.jsx("p",{style:{fontSize:"14px",color:"#94a3b8",maxWidth:"400px"},children:((a=this.state.error)==null?void 0:a.message)||"An unexpected error occurred."}),n.jsx("button",{onClick:()=>window.location.reload(),style:{marginTop:"16px",padding:"10px 28px",background:"hsl(var(--primary))",color:"#0a0a0f",border:"none",borderRadius:"8px",cursor:"pointer",fontSize:"14px"},children:"Refresh page"})]}):this.props.children}}const jc={title:"Quantum Dashboard",subtitle:"Monitor your QPU usage & test quantum functions",qpuActive:"QPU Active",tokensUsed:"Tokens Used",tokenLimit:"Token Limit",apiCalls:"API Calls",usage:"Usage",tokenUsage:"Token Usage",qrng:{title:"QRNG â Quantum Random",description:"Generate true random numbers from 72-qubit QPU",button:"Generate Random Number (0-1000)",result:"72-qubit QRNG result",success:"QRNG executed successfully",error:"QRNG failed"},qaoa:{title:"QAOA â Optimization",description:"Run portfolio optimization on quantum hardware",button:"Optimize Portfolio Allocation",qpu:"QPU",classical:"Classical",success:"QAOA executed on QPU",fallback:"QAOA classical fallback",error:"QAOA failed"},quickActions:"Quick Actions",apiDocs:"API Documentation",upgradePlan:"Upgrade Plan",mainDashboard:"Main Dashboard"},Gc={badge:"Welcome to Quantum",title:"Setup Your QPU",subtitle:"3 steps to quantum-powered inference",steps:{apiKey:"API Key",testQRNG:"Test QRNG",firstQAOA:"First QAOA",ready:"Ready!"},step1:{title:"Your API Key",description:"Your Quantum API key has been generated and is available in your dashboard.",findKey:"Find your full key in",dashboardLink:"Dashboard â API Keys",continue:"Continue"},step2:{title:"Test QRNG",description:"Let's generate your first quantum random number from the 72-qubit QPU.",button:"Generate QRNG",success:"QRNG test passed!",result:"Quantum random generated",error:"QRNG test failed"},step3:{title:"First QAOA Optimization",description:"Run a portfolio optimization using QAOA on quantum hardware.",button:"Run QAOA",success:"QAOA optimization complete!",result:"QAOA optimization complete",error:"QAOA test failed"},step4:{title:"You're Ready! ð",description:"Your quantum environment is set up. Start building with the full API.",goToDashboard:"Go to Dashboard",readDocs:"Read API Docs"}},Bc={companies:"Companies",tokensProcessed:"Tokens processed",compliant:"Compliant",apiOperational:"API operational"},Uc={signedUp:"{{name}} from {{location}} just signed up",upgraded:"{{name}} from {{location}} upgraded to Pro",saved:"{{name}} saved {{amount}} this month",timeAgo:"{{time}} min ago"},Qc={title:"Wait! ð",description:"Get 50K FREE tokens to test our ultra-fast API. No credit card required.",exclusiveOffer:"Exclusive offer",noCommitment:"100% free ⢠No commitment",claimTokens:"Claim my free tokens",trustNote:"Joined by 500+ companies ⢠OpenAI SDK compatible"},Vc={step1:"Sign up",step2:"Choose your plan or bundle",step3:"Pay for your plan",step4:"Get your API key on your Dashboard",dot1:"Sign-up",dot2:"Plan",dot3:"Payment",dot4:"API Key"},Hc={badge:"âï¸ COMPETITIVE ANALYSIS",title:"How We",titleHighlight:"Compare",subt
98itle:"See how Infernex stacks up against major inference providers across key metrics.",price:"Price",latency:"Latency",migration:"Ease of Migration",flexibility:"Model Flexibility",support:"Support Quality",anyscaleNote:"Strong open-source focus but limited model variety",clickToReset:"Click to reset",cta:"See Full Comparison",groqNote:"Fast but limited model selection",infernexDesc:"Best overall value across all dimensions",leader:"Leader",nvidiaNote:"Excellent hardware but premium pricing",togetherNote:"Good for open-source models, higher pricing"},Wc={badge:"Enterprise Reliability",title:"Better SLA than",titleHighlight:"OpenAI Direct",subtitle:"Multi-provider redundancy means higher uptime than any single provider can offer.",provider:"Provider",sla:"SLA Guarantee",failover:"Auto Failover",noPublicSla:"No public SLA",limited:"Limited",no:"No",multiProvider:"Multi-provider auto",keyMessage:"Better SLA than OpenAI. 95% cheaper.",explanation:"Our multi-provider architecture ensures <100ms failover. If one provider fails, your requests are automatically routed to the next available provider."},Kc={badge:"Intelligent Routing",title:"One API.",titleHighlight:"All Models.",subtitle:"Infernex routes your requests to the optimal provider, maximizing performance while minimizing costs.",providers:"LLM Providers",clients:"Your Business",client1:"Developers",client2:"Startups",client3:"Enterprise",savings:"savings",avgCost:"avg. cost",perMillion:"Per 1M tokens",feature1:"Intelligent Routing",feature2:"Auto Failover",feature3:"Cost Optimization",stat1:"Models Available",stat2:"Avg. Latency",stat3:"Uptime SLA",stat4:"Min. Latency",hubCentral:"Central Hub"},$c={playground:"Playground",calculate:"Compare",pricing:"Pricing",try:"Try"},Yc={backToHome:"Back to Home",backToBlog:"Back to Blog",back:"Back",cancel:"Cancel",error:"An error occurred",loading:"Loading...",next:"Next",skip:"Skip",submit:"Submit",success:"Success !",confirm:"Confirm",save:"Save",delete:"Delete",edit:"Edit",close:"Close",previous:"Previous",search:"Search",filter:"Filter",export:"Export",import:"Import",yes:"Yes",no:"No",enterprise:"Enterprise",checkoutError:"Checkout error",paymentError:"Payment error",loginFirst:"Please log in first",errorGeneric:"Error",live:"LIVE",enums:{high:"High",medium:"Medium",low:"Low",critical:"Critical",pending:"Pending",approved:"Approved",rejected:"Rejected",delayed:"Delayed",executing:"Executing",done:"Done",running:"Running",open:"Open",resolved:"Resolved",active:"Active",deployed:"Deployed",complex:"Complex",simple:"Simple",moderate:"Moderate",warning:"Warning",info:"Info",failed:"Failed"}},Zc={seo:{title:"Serverless AI Inference | Scale LLMs Without Infrastructure",description:"Deploy AI models without managing servers. Infernex serverless inference: <10ms latency, pay-per-token, auto-scaling, OpenAI-compatible. From $0.05/M tokens."},hero:{badge:"âï¸ SERVERLESS INFERENCE",title:"Scale LLMs",titleHighlight:"Without Infrastructure",subtitle:"Zero servers. Zero GPUs. Zero cold starts. Deploy AI inference that scales from 0 to millions of requests â pay only for what you use.",ctaStart:"Start Free",ctaPrice:"View Pricing"},features:{title:"Why Serverless with",titleHighlight:"Infernex",zeroInfra:{title:"Zero Infrastructure",desc:"No servers to manage, no GPUs to provision. Send requests to our API and we handle the rest."},coldStart:{title:"<10ms Cold Start",desc:"Unlike AWS Lambda or Google Cloud Functions, our inference endpoints are always warm. No cold start penalty."},payPerToken:{title:"Pay Per Token",desc:"No idle costs. Pay only for the tokens you consume. From $0.05 per million tokens."},autoScaling:{title:"Auto-Scaling",desc:"Handle 10 or 10 million requests per day without configuration changes. Scales instantly."},uptime:{title:"99.9% Uptime SLA",desc:"Enterprise-grade reliability with automatic failover across multiple regions."},openai:{title:"OpenAI Drop-in",desc:"Change one line of code. Our API is 100% compatible with OpenAI SDK. Zero migration friction."}},useCases:{title:"Use Cases",chatbots:{title:"Chatbots & Assistants",desc:"Deploy conversational AI without managing GPU clusters. Scale from 0 to millions of messages."},rag:{title:"RAG Pipelines",desc:"Embed retrieval-augmented generation into your app. Low-latency inference for real-time search."},batch:{title:"Batch Processing",desc:"Process millions of documents overnight. Pay only for what you use, no idle GPU costs."},edge:{title:"Edge AI",desc:"Serve inference from the closest region. Sub-10ms latency for time-sensitive applications."}},code:{title:"Deploy in",titleHighlight:"3 Lines"},faq:{title:"FAQ",q1:{question:"What is serverless AI inference?",answer:"Serverless AI inference means you don't manage any infrastru
98cture. You send API requests and receive responses. We handle GPU provisioning, scaling, load balancing, and failover automatically."},q2:{question:"How does pricing compare to self-hosted?",answer:"Self-hosted GPU instances cost $2-10/hour whether you use them or not. With Infernex serverless, you pay per token â typically 60-95% cheaper for variable workloads."},q3:{question:"Is there a cold start problem?",answer:"No. Unlike generic serverless platforms, our inference endpoints are always warm. First-request latency is identical to subsequent requests."},q4:{question:"Can I use my existing OpenAI code?",answer:"Yes. Change the base URL and API key â that's it. Our API is fully compatible with the OpenAI SDK, LangChain, LlamaIndex, and all major frameworks."}},cta:{title:"Ready to Go Serverless?",subtitle:"Start with 1M free tokens. No credit card required.",button:"Get Started Free"}},Jc={seo:{title:"Hire Quant Alternative | Quantum Computing for Finance 2026",description:"Looking to hire a quant? Our quantum-accelerated inference API outperforms traditional quants in high-frequency financial modeling. 72-qubit powered. From $1,999/mo."},hero:{badge:"ð§ QUANTUM AI vs HUMAN QUANT",title:"Stop Hiring",crossedOut:"$200k Quants",titleHighlight:"Deploy Quantum AI",subtitle:"72-qubit quantum computing. Sub-50ms decisions. Zero human error. Infernex QPU replaces your next quant hire â at a fraction of the cost.",ctaQpu:"Explore QPU Plans",ctaComparison:"Full Comparison"},stats:{qubits:"Qubits",savings:"Cost Savings",latency:"Decision Latency",availability:"Availability"},comparison:{title:"Quant Analyst vs",titleHighlight:"Infernex QPU",criteria:"Criteria",humanQuant:"Human Quant",infernexQpu:"Infernex QPU",annualCost:"Annual Cost",hiringTime:"Hiring Time",availability:"Availability",scalability:"Scalability",humanError:"Human Error",quantumComputing:"Quantum Computing",modelsProcessed:"Models Processed",decisionLatency:"Decision Latency",cognitiveBias:"Cognitive Bias",vacationSick:"Vacation/Sick Days"},useCases:{title:"What Your QPU",titleHighlight:"Replaces",portfolioOpt:{title:"Portfolio Optimization",desc:"QAOA-powered asset allocation across unlimited portfolios simultaneously."},riskModeling:{title:"Risk Modeling",desc:"Real-time VaR, CVaR, and stress testing with quantum random sampling."},pricing:{title:"Pricing Models",desc:"Options pricing, derivatives valuation with quantum advantage."},trading:{title:"Algorithmic Trading",desc:"Sub-millisecond signal generation and execution."},analytics:{title:"Client Analytics",desc:"Behavioral analysis and churn prediction at scale."},monteCarlo:{title:"Quantum Monte Carlo",desc:"QRNG-powered simulations with true quantum randomness."}},faq:{title:"Common",titleHighlight:"Questions",q1:{question:"Can AI really replace a quantitative analyst?",answer:"For many tasks, yes. Infernex QPU handles portfolio optimization, risk modeling, and algorithmic trading with 72-qubit quantum computing â 24/7, with zero human error and sub-50ms latency."},q2:{question:"What about the human judgment a quant provides?",answer:"Infernex augments human judgment rather than replacing it entirely. Your team focuses on strategy while Infernex handles computation, data processing, and model execution at quantum speed."},q3:{question:"How much cheaper is Infernex vs hiring a quant?",answer:"A junior quant costs $125kâ$150k/year. A senior quant costs $200kâ$300k. Infernex Quantum Pro starts at $1,999/month ($23,988/year) â up to 92% savings with superior capabilities."},q4:{question:"Is quantum computing actually used?",answer:"Yes. Infernex Quantum uses a 72 superconducting qubit processor for QAOA optimization and QRNG random number generation â real quantum hardware, not simulation."}},cta:{title:"Your Next Quant Hire Costs $1,999/month",subtitle:"72 qubits. Zero bias. Infinite scale. Start today.",ctaQpu:"See QPU Plans",ctaComparison:"Detailed Comparison"}},Xc={seo:{title:"Enterprise AI Cost Reduction 2026 â Save 95% on Inference",description:"Enterprise AI inference at 95% lower cost. SOC 2 compliant, 99.99% SLA, sub-10ms latency, 72-qubit quantum computing. Migrate in 2 weeks. Get a quote."},hero:{badge:"ð¢ ENTERPRISE AI INFRASTRUCTURE",title:"Cut Enterprise AI Costs by",titleHighlight:"95%",subtitle:"Same models. Same quality. SOC 2 compliant. 99.99% SLA. Sub-10ms latency. 72-qubit quantum computing. Trusted by leading financial institutions.",ctaDemo:"Request Enterprise Demo",ctaRoi:"Calculate ROI"},stats:{savings:"Cost Savings",latency:"Latency",uptime:"Uptime SLA",qubits:"Quantum Qubits"},benefits:{title:"Enterprise-Grade",titleHighlight:"AI Infrastru
98cture",costReduction:{title:"95% Cost Reduction",desc:"Cut annual AI spending from $5M+ to under $250K with drop-in API migration. No code changes."},lowLatency:{title:"Sub-10ms Latency",desc:"Faster than OpenAI, Anthropic, and Google. Critical for real-time trading, fraud detection, and customer-facing AI."},sla:{title:"99.99% Uptime SLA",desc:"Enterprise-grade reliability with contractual SLA. Higher than AWS Bedrock (99.9%) and Azure OpenAI (99.9%)."},compliance:{title:"SOC 2 & GDPR Compliant",desc:"Data stays in Switzerland. Full audit trail, encryption at rest, and zero data retention policy."},infrastructure:{title:"Dedicated Infrastructure",desc:"Isolated compute for enterprise clients. No noisy neighbors. Custom rate limits and priority queuing."},quantum:{title:"72-Qubit Quantum Computing",desc:"Infernex Quantum QPU access for QAOA optimization and QRNG â unique competitive advantage."}},caseStudies:{title:"Enterprise",titleHighlight:"ROI Examples",before:"Before",after:"After",saved:"saved"},migration:{title:"Enterprise",titleHighlight:"Migration Path",discovery:{title:"Discovery",desc:"We audit your current AI stack and spending"},poc:{title:"POC",desc:"2-week proof of concept with your workloads"},migration:{title:"Migration",desc:"Gradual traffic shift with zero downtime"},optimize:{title:"Optimize",desc:"Quantum-optimized inference at 95% lower cost"}},faq:{title:"Enterprise",titleHighlight:"FAQ",q1:{question:"How much can enterprise customers save?",answer:"Enterprise customers typically save 90-97% on AI inference costs. A company spending $5M/year on OpenAI can reduce to under $250K with Infernex â same models, same quality, fraction of the cost."},q2:{question:"Is there a minimum commitment?",answer:"Enterprise plans start at custom pricing based on volume. We offer flexible monthly billing with volume discounts. No long-term lock-in required."},q3:{question:"How does the migration work for enterprise?",answer:"Our enterprise team handles migration end-to-end. Typical migration takes 2-4 weeks including testing, validation, and gradual traffic shift. Zero downtime guaranteed."},q4:{question:"What about data security and compliance?",answer:"Infernex is SOC 2 Type II compliant, GDPR compliant, with data hosted in Switzerland. We offer zero data retention, encrypted communications, and full audit logging."},q5:{question:"Do you offer quantum computing for enterprise?",answer:"Yes. Enterprise QPU plans include access to Infernex Quantum's 72-qubit processor for QAOA optimization and QRNG. This is unique to Infernex â no other inference provider offers quantum acceleration."}},cta:{title:"Ready to Cut Your Enterprise AI Budget by 95%?",subtitle:"Custom pricing. Dedicated support. SOC 2 compliant. Zero downtime migration.",ctaDemo:"Request Enterprise Demo",ctaComparison:"See Comparison vs OpenAI"},brand:{os:"Enterprise OS",tagline:"Infernex Enterprise Operating System"},live:{label:"LIVE DATA",lastSync:"Last sync"},sidebar:{home:"Home",aiWorkforce:"AI Workforce",workforceMarketplace:"AI Marketplace",workforceDashboard:"Workforce Dashboard",workforceConnectors:"Connectors",workforceRouter:"AI Router",workforceSafety:"Safety Layer",founder:"Founder Center",investor:"Investor Room",crm:"CRM",enterpriseAccounts:"Enterprise Accounts",partners:"Partner Center",marketplace:"Marketplace Center",finops:"FinOps",developer:"Developer Console",dataRoom:"Data Room",dueDiligence:"Due Diligence",documents:"Documents",pitch:"Pitch Deck",analytics:"Analytics",copilot:"AI Copilot",settings:"Settings",bizdev:"Business Dev",aiSupervisor:"AI Supervisor",roiEngine:"ROI Engine",automation:"Automation Center",directories:"Directory Submissions",knowledge:"Knowledge Center",knowledgeGraph:"Knowledge Graph",optimizer:"Global Optimizer",executive:"Executive Insights",security:"Security Center",partnerPortal:"Partner Portal",customerPortal:"Customer Portal",verification:"System Verification",auditLog:"Audit Log",aiGovernance:"AI Governance",aiPerformance:"AI Performance",firecrawlHealth:"Firecrawl Health",emailMonitor:"Email Monitoring",deliverability:"Deliverability",campaignPreview:"Campaign Preview",prospectTargets:"Prospect Targets",referrals:"Referrals",partnershipRoadmap:"Partnership Roadmap",outreachDashboard:"Outreach & Quota",warRoom:"War Room YC",superAdmin:"Super Admin",companyBuilder:"Company Builder"},warRoom:{title:"â¡ War Room YC",subtitle:"Live SLO monitoring",loading:"Loading War Roomâ¦"}
98,supervisor:{title:"AI Supervisor",subtitle:"Live orchestration of the 500,000 MiroFish agents powering every module.",connected:"Connected agents",working:"Working",idle:"Idle",avgLatency:"Avg latency",tasks:"Tasks completed",predictions:"Predictions",docs:"Documents generated",sims:"Financial simulations",queue:"Queue depth",gpu:"GPU usage",memory:"Memory usage",routingTitle:"Task routing",agents:"agents",routing:{financial:"Financial Agents",sales:"Sales Agents",investor:"Investor Agents",documents:"Documentation Agents",marketplace:"Marketplace Agents",security:"Security Agents",planning:"Planning Agents",roi:"ROI Agents"}},roi:{title:"ROI Engine",subtitle:"Enterprise-grade scenario simulator connected to the MiroFish intelligence layer.",variables:"Variables",enterprises:"Enterprise customers",maus:"Monthly active customers",tokensM:"Token usage (millions/mo)",price:"Price per million tokens",gpu:"GPU cost per million tokens",opex:"Operational expenses",partner:"Partner commission",cac:"Customer acquisition cost",forecast12:"12-month forecast",sensitivity:"EBITDA sensitivity",aiAssessment:"MiroFish AI assessment",runAI:"Ru
98n AI assessment",analyzing:"Analyzingâ¦",assessmentEmpty:"Run the AI assessment to see optimizations and risks.",metric:{monthlyRevenue:"Monthly revenue",arr:"ARR",grossProfit:"Gross profit",ebitda:"EBITDA (yr)",ltv:"LTV",payback:"Payback"}},automation:{title:"Automation Center",subtitle:"No-code workflows orchestrated by the MiroFish agent network.",templates:"Templates",use:"Use template",custom:"Custom workflow",placeholder:"Workflow name",create:"Create",created:"Workflow created",active:"Active workflows",empty:"No workflows yet.",run:"Run",executed:"Workflow executed",logs:"Recent runs",noRuns:"No runs yet."},knowledge:{title:"AI Knowledge Center",subtitle:"Unified answers across products, pricing, roadmap, partners, security, API and more.",placeholder:"Ask anything about Infernexâ¦",search:"Search",answer:"Answer",domain:{all:"All",products:"Products",pricing:"Pricing",roadmap:"Roadmap",partners:"Partners",marketplace:"Marketplace",docs:"Documentation",investors:"Investors",strategy:"Strategy",sales:"Sales",security:"Security",architecture:"Architecture",api:"API",sdk:"SDK"}},copilotPage:{title:"AI Copilot",subtitle:"Founder and Investor copilots powered by the 500,000 MiroFish agents.",quick:"Quick prompts",empty:"Pick a quick prompt or ask anything.",thinking:"MiroFish is thinkingâ¦",placeholder:"Ask a questionâ¦",persona:{founder:"Founder",investor:"Investor"}},analyticsPage:{title:"Predictive Analytics",subtitle:"Best / expected / worst case forecasts generated by the MiroFish agents.",forecast12:"12-month scenarios",runAI:"Generate AI analysis",aiAnalysis:"AI analysis",dim:{revenue:"Revenue",customers:"Customers",arr:"ARR",cashflow:"Cash flow",infrastructure:"Infrastructure",marketplace:"Marketplace adoption",partners:"Partners",hiring:"Hiring",risk:"Risk"}},founder:{title:"Founder Dashboard",subtitle:"Real-time view of every signal that matters.",infraStatus:"Infrastructure Status",tokenRouting:"Token Routing"},investor:{seoTitle:"Investor Portal · Infernex",welcomeTitle:"Welcome to the Infernex Investor Portal",welcomeBody:"This portal is fully automated. Every KPI is synchronized, every document regenerated from live company data. No manual reports, nothing outdated.",point1:"Live financial synchronization",point2:"Automated document regeneration",point3:"Verified data sources",point4:"Last updated in real time",confidence:"Investor Confidence",components:{financial:"Financial",roadmap:"Roadmap",pitch:"Pitch",dataRoom:"Data Room",partners:"Partners",marketplace:"Marketplace",security:"Security",technical:"Technical"},runDD:"Run AI Due Diligence"},kpi:{mrr:"MRR",arr:"ARR",founders:"Founders Club",enterpriseCustomers:"Enterprise Customers",customers:"Customers",activeSubs:"Active Subscriptions",geoActive:"GEO Active",profit30:"Profit 30d",runway:"Runway",runwayHint:"Profitable since day one",live:"Synchronized"},notifications:{title:"Notifications",empty:"No notifications yet.",markAllRead:"Mark all read"},documents:{title:"Document Engine",subtitle:"Live, regenerated from company data â PDF, DOCX, PPTX, XLSX.",generate:"Generate document",recent:"Recent documents",generated:"Document generated",error:"Generation failed",none:"No documents yet.",types:{pitch_deck:"Pitch Deck",executive_summary:"Executive Summary",business_plan:"Business Plan",white_paper:"White Paper",financial_model:"Financial Model",cap_table:"Cap Table",roadmap:"Roadmap",roi_report:"ROI Report",partner_agreement:"Partner Agreement",nda:"NDA",sales_deck:"Sales Deck",one_pager:"One Pager",architecture:"Architecture Overview",security_doc:"Security Documentation",sdk_doc:"SDK Documentation",api_doc:"API Documentation"}},copilot:{open:"Open AI Copilot",title:"AI Copilot",welcome:"Ask me anything: MRR, generate pitch, roadmap progressâ¦",placeholder:"Ask the Copilotâ¦",thinking:"Thinkingâ¦",error:"The Copilot is unavailable right now."},stub:{comingSoon:"Module ready â full UI ships in Prompt #2.",crm:"Pipeline, deals and enterprise contacts.",enterpriseAccounts:"Manage your Fortune-500 accounts.",partners:"Reseller and channel partner workspace.",marketplace:"AWS, Azure and GCP marketplace listings.",finops:"Margins, cost analytics and unit economics.",developer:"API keys, webhooks, SDK and logs.",dataRoom:"Investor-grade encrypted data room.",pitch:"Live pitch deck pulled from real KPIs.",analytics:"Cross-domain analytics and exports.",copilot:"Standalone copilot workspace.",settings:"Worksp
98ace, roles and integrations."},common:{search:"Searchâ¦",empty:"No data yet",created:"Created",create:"Create"},mirofish:{title:"MiroFish Intelligence Network",agents:"500,000 AI Agents Connected",live:"LIVE",lastSync:"Last sync",nextSync:"Next sync",orchestration:"Orchestration active",verified:"Signed & verified",modules:{orchestration:"Orchestration",optimization:"Optimization",prediction:"Prediction",knowledge:"Knowledge",learning:"Learning",document:"Documents",investment:"Investment",financial:"Financial",risk:"Risk"}},ai:{verified:"Verified",confidence:"Confidence",sync:"Sync",generatedBy:"Generated by",ts:"At"},dueDiligence:{seoTitle:"AI Due Diligence â Infernex Investor Portal",title:"AI Due Diligence",subtitle:"Institutional-grade due diligence generated by the MiroFish network from live company data.",runCta:"Run AI Due Diligence",investmentScore:"AI Investment Score",download:"Download report",dims:{financial:"Financial",growth:"Growth",revenueQuality:"Revenue quality",infra:"Infrastructure",marketplace:"Marketplace",ecosystem:"Ecosystem",product:"Product",docs:"Documentation",security:"Security",techDebt:"Tech debt",customer:"Customer growth",retention:"Retention",margin:"Gross margin",scalability:"Scalability",aiReadiness:"AI readiness"}},dataRoom:{seoTitle:"Data Room â Infernex Investor Portal",title:"Virtual Data Room",subtitle:"Encrypted, versioned, watermarked â every document regenerated from live company data.",empty:"No documents in this folder yet. Generate one below.",autoGenerate:"Auto-generate from live data",regenerate:"Regenerate",generated:"Document generated",verified:"MiroFish verified",folders:{corporate:"Corporate",financial:"Financial",technology:"Technology",security:"Security",legal:"Legal",sales:"Sales",marketing:"Marketing",operations:"Operations",product:"Product",marketplace:"Marketplace",partners:"Partners",investors:"Investors",hr:"HR"}},pitch:{seoTitle:"AI Pitch Deck â Infernex Investor Portal",title:"AI Pitch Deck",subtitle:"Auto-generated pre-seed investor deck â every slide synchronized with live operations and Prompt 4 projections.",badge:"Investor Deck · Auto-synced",regenerate:"Regenerate narrative",regenerated:"Pitch deck regenerated",present:"Present",slides:{cover:{title:"Hire AI employees and create your company in one click.",body:"Infernex is the AI Workforce platform: you hire autonomous AI employees who build and run your company 24/7 â brand, website, legal, finance, sales, growth. Our own inference engine, the Infernex Core engine, is unique in the world â it is the foundation Infernex AI is built on, and the reason we can cut inference costs by 95% compared to other LLMs."},vision:{title:"Vision",body:"Every company becomes AI-native: teams of digital employees that own outcomes. Infernex is the operating system that hires, governs and pays them â powered by our own inference engine so the cost of intelligence never limits the business."},problem:{title:"The Problem",body:"Hiring is slow and expensive, SaaS seats pile up, BPO doesn't scale, and generic AI tools produce text instead of results. Nobody owns the outcome â and inference bills make automation unpredictable."},solution:{title:"The Solution",body:"Infernex AI Workforce: hire AI employees (Sales, Support, Finance, Legal, Ops, Growth) supervised by a Master AI Agent that runs the company. Everything runs on our own inference engine â same models, up to 95% cheaper â so margins stay structurally superior."},technology:{title:"Technology",body:"Workforce layer: agent runtime, role templates, memory, approval engine, multi-tenant RLS, 93 enterprise connectors. Engine layer: Infernex Core Router + photonic ONN + 72-qubit QPU + TSU PBit (probabilistic tensor compute) + MiroFish orchestration (a 500,000 AI-agent consensus) â selecting the cheapest capable model for every task an AI employee performs."},ai:{title:"AI Architecture",body:"Master AI Agent supervises specialized AI employees; each task is routed by the Core Router to the cheapest capable model. Themis governance, audit trail and human approval gates make autonomy safe for enterprises."},product:{title:"Product",body:"One platform, one journey: AI Employee Marketplace (25+ roles) â Company Builder â Master AI Agent (CEO/Director mode) â Governance, ROI and FinOps centers. The Inference API and SDKs remain available for developers, but as the engine beneath the workforce."},market:{title:"Market",body:"We sell labor, not tokens. Target: the digital labor and business-services market (SMB, mid-market, enterprise) replacing SaaS seats, agencies and BPO. Combined workforce + inference TAM above $300B by 2028."},competition:{title:"Competition",body:"OpenAI/Anthropic/Groq sell tokens. Salesforce/ServiceNow sell seats. Agent startups sell frameworks. Infernex sells staffed outcomes â AI employees that run a whole company, on an inference engine we own and price 95% below the market."},advantages:{title:"Competitive Advantages",body:"Vertical integration: workforce on top of our own inference engine · up to 95% lower cost per task · 25+ ready-to-hire AI roles · 93 connectors · Master AI Agent orchestration · Swiss sovereignty & governance · self-service onboarding · Founders Club (500 seats)."},customers:{title:"Customers",body:"Founders creating a company from an idea · SMBs replacing missing headcount · mid-market teams automating Sales, Support, Finance and Ops · enterprises deploying governed AI staff. Developers still use the inference API â it is the entry point, not the destination."},traction:{title:"Traction",body:"Live KPIs â Founders Club seats, MRR/ARR, customers and AI-employee activity synchronized every 30 seconds."},businessModel:{title:"Business Model",body:"We charge for AI employees, with inference tokens included in every plan. Starter 59â¬/mo (Founders Club lifetime, 500 seats), Business, Enterprise custom â priced per AI employee. Inference usage bundles and marketplace revenue-share are secondary streams with structurally high margin."},gtm:{title:"Go-To-Market",body:"PLG: create your company and hire your first AI employee free â Founders Club 59â¬/mo lifetime â Marketplace self-service upsell of additional AI employees â VC Perks, partner channel and cloud marketplaces â Global 2000 outbound via our own AI workforce."},marketplace:{title:"Marketplace Strategy",body:"AWS, Azure, GCP and B2B directories stay in standby until manual activation; exports, lead sync and listings are ready."},partnerships:{title:"Strategic Partnerships",body:"Partner program with 10% recurring MRR commission, VC Perks outreach and automated partner lead tracking."},finops:{title:"FinOps",body:"Cost per AI employee, quotas, hard caps and margin visibility. Because we own the inference engine, gross margin per employee stays above 80%."},roi:{title:"ROI",body:"3-year plan â Raise $1M pre-seed to cover: AI workforce & growth ops $432k · CEO 5â10k CHF/mo + Swiss social charges ~$333k · Swiss fiduciary ~$24k · SaaS/infra $108k · marketplaces + SOC2/ISO27001 $60k · buffer $43k. ROI forecast: Y1 ARR ~180k⬠(Founders Club + early paid) · Y2 ~850k⬠(marketplaces + partners) · Y3 ~2.6M⬠(enterprise + sovereign). Break-even Q3 Y2 · 3à on invested capital by end of Y3."},financials:{title:"Pre-seed Financial Projection",body:"Prompt 4 assumptions included: 500 Founder seats à 59â¬/month = 29,500⬠gross MRR when full; 500K trial tokens; paid 5M-token bundles/overage; target gross margin above 80% with paid modules kept in standby until activation."},roadmap:{title:"Roadmap",body:"Phase 1: Founders Club and self-service API. Phase 2: VC Perks and partner program. Phase 3: marketplaces. Phase 4: enterprise/sovereign scale â paid tools remain standby until approved."},team:{title:"Team",body:"Ricardo Cadet-Marthe, Founder & CEO (Geneva, Switzerland) â ready to relocate to the USA or UAE. CEO salary 5â10k CHF/mo. Infernex runs on its own product: the Master AI Agent supervises a full workforce of AI employees (growth, marketing, sales, SEO, content, support, RevOps, analytics) â the same solution we sell to company builders. Swiss fiduciary firm handles accounting, VAT, payroll and year-end closing."},investment:{title:"Investment",body:"Pre-seed raise: $1M / ~850k⬠to cover 3 years. Use of funds: AI workforce & growth ops $432k · CEO salary + Swiss social charges ~$333k · Swiss fiduciary ~$24k · SaaS & infra $108k · marketplaces + SOC2/ISO27001 $60k · buffer $43k. ROI forecast â Y1 ARR ~180k⬠(Founders Club 500 à 59⬠+ early paid) · Y2 ~850k⬠(marketplaces + partners live) · Y3 ~2.6M⬠(enterprise + sovereign). Break-even Q3 Y2, 3à return on invested capital by end of Y3."},closing:{title:"Join Us",body:"Infernex â hire AI employees, create your company in one click. Digital labor on a sovereign, 95%-cheaper inference engine. Swiss-made."},investorMetrics:{title:"Investor metrics vs YC targets",body:"Live indicators auto-synced from api_request_logs â DAU, MAU, latency, success rate, models available and requests per active user, side-by-side with the benchmarks Y Combinator expects."},whyWorkforce:{title:"Why AI employees, not chatbots",body:"Companies don't need more prompts â they need staff. An Infernex AI employee has a role, objectives, tools, permissions and KPIs: qualified pipeline, resolved tickets, closed books, drafted contracts. Deploy in minutes, measure ROI in days."},teamVision:{title:"Team & Vision",teamTitle:"Team",teamBody:"Ricardo Cadet-Marthe â CEO & Founder (Geneva ð¨ð), ready to relocate USA/UAE. Master AI Agent â runs Infernex itself with a full team of AI employees (growth, marketing, sales, SEO, support, RevOps). Swiss fiduciary â accounting, VAT, payroll, year-end.",visionTitle:"Vision",visionBody:"Become the operating system of AI Workforce for SMB, mid-market and enterprise across the EU, US, MENA and APAC."},creationFlow:{title:"From idea to running company â 6 steps",body:"1) Describe your idea. 2) Infernex generates the blueprint: legal form, country, budget, KPIs. 3) Hire your AI employees from 25+ roles. 4) Connect your stack (93 connectors). 5) One-click deploy: brand, domain, website, cloud, compliance. 6) The Master AI Agent runs operations, growth and finance 24/7. Human approval on every irreversible or spending action."}}},crm:{title:"CRM",subtitle:"Enterprise pipeline & accounts",seo:{title:"CRM â Infernex Enterprise OS",desc:"Enterprise CRM"},addCompany:"Add company",tabs:{companies:"Companies",pipeline:"Pipeline"},col:{company:"Company",country:"Country",industry:"Industry",stage:"Status",health:"Health",savings:"Annual savings",website:"Website"},stages:{lead:"Lead",qualified:"Qualified",discovery:"Discovery",demo:"Demo",validation:"Validation",pilot:"Pilot",marketplace:"Marketplace",security:"Security",legal:"Legal",procurement:"Procurement",negotiation:"Negotiation",signed:"Signed",won:"Won",lost:"Lost"},actions:{superAdmin:"Super Admin",companyBuilder:"Company Builder"}},accounts:{title:"Enterprise Accounts",subtitle:"Fortune-class customers",plan:"Plan",started:"Started"},partners:{seo:{title:"Partner Center â Infernex",desc:"Manage cloud, FinOps, consulting, referral, VC and integrator partners."},title:"Partner Center",subtitle:"Cloud, FinOps, consulting, referral, VC and integrator partners.",kpi:{total:"Partners",mrr:"MRR generated",arr:"ARR generated",due:"Commissions due"},col:{name:"Partner",type:"Type",tier:"Tier",country:"Country",mrr:"MRR",due:"Due",status:"Status"}},marketplace:{seo:{title:"Marketplace Center â Infernex",desc:"AWS, Azure and GCP marketplace listings, checklists and progress."}
98,title:"Marketplace Center",subtitle:"AWS, Azure and GCP listings with automated checklist progress.",progress:"Progress"},finops:{seo:{title:"FinOps Center â Infernex",desc:"FinOps partners, integrations and cloud/AI spend optimization."},title:"FinOps Center",subtitle:"Track FinOps partners, integrations and cumulative AI savings.",kpi:{total:"Partners",active:"Active",savings:"Savings YTD",avgShare:"Avg. revenue share"},col:{name:"Partner",priority:"Priority",integration:"Integration",api:"API",revShare:"Rev. share",website:"Website"}},bizdev:{seo:{title:"Business Development â Infernex",desc:"Global BizDev pipeline with AI scoring and regional segmentation."},title:"Business Development",subtitle:"Global pipeline with AI scoring, ROI estimates and regional segmentation.",addTarget:"New target",kpi:{total:"Targets",spend:"AI spend is.",savings:"Savings potential",avgScore:"Avg. AI score"},col:{company:"Company",region:"Region",country:"Country",industry:"Industry",priority:"Priority",spend:"AI spend",savings:"Savings",score:"AI score",status:"Status"},regions:{all:"All",na:"North America",eu:"Europe",asia:"Asia",me:"Middle East",sa:"South America",africa:"Africa",oceania:"Oceania"}},developer:{title:"Developer Console",subtitle:"Explore APIs, inspect logs, manage webhooks & SDKs â routed through the MiroFish gateway.",kpi:{keys:"API Keys",avgLatency:"Avg Latency",errors:"4xx / 5xx",incidents:"Open Incidents"},tabs:{explorer:"API Explorer",logs:"Live Logs",webhooks:"Webhooks",sdk:"SDK Manager"},explorer:{title:"API Explorer",send:"Send request"},logs:{empty:"No recent requests."},webhooks:{title:"Registered webhooks"},sdk:{title:"Official SDKs"},copy:"Copy",copied:"Copied to clipboard"},security:{title:"Security Center",subtitle:"Real-time monitoring of RLS, encryption, sessions, keys and compliance.",kpi:{admins:"Admins",apiKeys:"API Keys",critical:"Critical Events",tickets:"Open Tickets"},compliance:{title:"Compliance status",rls:"Row-Level Security",encryption:"AES-256 Encryption",backups:"Encrypted Backups",gdpr:"GDPR / DSR",csp:"Strict CSP",otp:"6-digit OTP"},events:{title:"Recent security events",empty:"No security events recorded."}},optimizer:{title:"Global Optimizer",subtitle:"Autonomous MiroFish recommendations across every business surface.",regenerate:"Regenerate",recommendations:"Prioritized recommendations",confidence:"confidence"},executive:{title:"Executive Insights",subtitle:"Daily AI-generated reports for every leadership role.",reports:{ceo:"CEO Report",cto:"CTO Report",sales:"Sales Report",finance:"Finance Report",investor:"Investor Report",marketplace:"Marketplace Report",partner:"Partner Report",security:"Security Report",risk:"Risk Report"},samples:{ceo:`## CEO Report 99 100- MRR trend: healthy, driven by Founders Club activations 101- Runway: extended, no cash risk 102- Next 30 days: focus on Enterprise POCs and marketplace listings`,cto:`## CTO Report 103 104- Infrastructure: nominal, edge cache hit rate 78% 105- Model routing: Flash-Lite 62% / Flash 31% / Pro 7% 106- Priorities: rotate keys, add regional failover`,sales:`## Sales Report 107 108- Pipeline: healthy inbound from GEO citations 109- POC conversion: 22% 110- Focus: expiring POCs & enterprise renewals`,finance:`## Finance Report 111 112- Gross margin: >85% 113- Payment infra: Stripe live, webhooks green 114- Recommendation: introduce annual Founders tier`,investor:`## Investor Report 115 116- Traction ahead of plan 117- MiroFish moat expanding 118- Data Room & Pitch Deck fully synchronized`,marketplace:`## Marketplace Report 119 120- 2 listings live, 3 in preparation 121- Focus on Vercel & Cloudflare marketplaces`,partner:`## Partner Report 122 123- Autonomous onboarding active 124- Top partners generating recurring commissions`,security:`## Security Report 125 126- RLS enforced on every public table 127- No critical events in 30 days
128- CSP strict, OTP hardening on`,risk:`## Risk Report 129 130- Supplier concentration: low 131- Regulatory posture: GDPR ready 132- Key rotation cadence: quarterly`},download:"Download",generating:"MiroFish network is generating the reportâ¦",empty:"Select a report to generate."},knowledgeGraph:{title:"Enterprise Knowledge Graph",subtitle:"Every entity of Infernex connected, searchable and MiroFish-aware.",search:"Search entities, typesâ¦"},partnerPortal:{title:"Partner Portal",subtitle:"Everything a partner needs: pipeline, commissions, assets, contracts.",kpi:{partners:"Partner accounts",referrals:"Referrals",commissions:"Commissions",assets:"Marketing assets"},invoices:"Recent invoices",marketingAssets:"Marketing assets",empty:"Nothing yet."},customerPortal:{title:"Enterprise Customer Portal",subtitle:"Live usage, savings, AI recommendations and support in one place.",kpi:{tokens:"Tokens used",cost:"Cost (30d)",saved:"Saved vs incumbents",tickets:"Open tickets"},recommendations:"MiroFish recommendations",recentTickets:"Recent tickets",noTickets:"No open tickets.",recos:{cache:"Enable semantic cache on RAG endpoints",batch:"Batch small requests via streaming SDK",tier:"Move steady traffic to the Founders tier"}},settingsPage:{title:"Settings",subtitle:"Preferences, notifications, security and MiroFish routing.",general:"General",security:"Security",language:"Language",notifications:"In-app notifications",autoOptimize:"Autonomous optimization",mirofishRouting:"MiroFish AI routing",passwordPolicy:"Password policy",rls:"Row-Level Security",encryption:"Data encryption"},verification:{title:"Production Verification Report",subtitle:"Automated audit of the Infernex Enterprise Operating System.",checks:{existingArchitecture:"Existing architecture preserved",existingDashboard:"Existing Dashboard preserved",nebulaTheme:"Nebula theme preserved",mirofishReused:"500,000 MiroFish agents reused",noDuplicatedAI:"No duplicated AI",modulesSynchronized:"All modules synchronized",investorPortal:"Investor Portal operational",partnerPortal:"Partner Portal operational",crm:"CRM operational",marketplace:"Marketplace Center operational",finops:"FinOps operational",aiCopilots:"AI Copilots operational",pitchDeck:"Pitch Deck Engine operational",dataRoom:"Data Room operational",roiEngine:"ROI Engine operational",documentGenerator:"Document Generator operational",downloads:"Downloads operational",i18n:"i18n operational (15 languages)",responsive:"Responsive design verified",productionReady:"Production ready"}}},ed={seo:{title:"Quantum-Accelerated AI Inference 2026 â 72-Qubit QAOA & QRNG",description:"Quantum-accelerated AI inference with 72-qubit Infernex Quantum. QAOA optimization, QRNG random generation. 1000x faster portfolio optimization. Start now."},hero:{badge:"âï¸ QUANTUM-ACCELERATED INFERENCE",title:"AI Inference Powered by",titleHighlight:"72-Qubit Quantum",subtitle:"The world's first AI inference API with integrated quantum computing. QAOA optimization. QRNG random generation. Infernex Quantum â 72 superconducting qubits.",ctaQpu:"Explore QPU Plans",ctaVsQuant:"QPU vs Human Quant"},stats:{qubits:"Qubits",speedup:"Speedup",latency:"Latency",perMonth:"/month"},features:{title:"Quantum",titleHighlight:"Capabilities",qubits:{title:"72 Superconducting Qubits",desc:"Infernex Quantum processor â one of the most powerful quantum computers accessible via API."},qaoa:{title:"QAOA Optimization",desc:"Quantum Approximate Optimization Algorithm for portfolio allocation, scheduling, and combinatorial problems."},qrng:{title:"QRNG Random Generation",desc:"True quantum random numbers for Monte Carlo simulations, cryptography, and unbiased A/B testing."},hybrid:{title:"Hybrid Classical-Quantum",desc:"Seamless fallback to classical algorithms. If QPU is unavailable, your workloads never stop."}
132,latency:{title:"Sub-50ms Decision Latency",desc:"Quantum-accelerated decisions faster than any human analyst or classical-only system."},security:{title:"Enterprise-Grade Security",desc:"SOC 2, GDPR compliant. Quantum key distribution ready. Swiss-hosted infrastructure."}},benchmarks:{title:"Quantum vs Classical",titleHighlight:"Benchmarks",task:"Task",classical:"Classical",quantum:"Quantum (72Q)",speedup:"Speedup"},useCases:{title:"Industry",titleHighlight:"Use Cases",faster:"faster"},code:{title:"3 Lines of",titleHighlight:"Code"},faq:{title:"Quantum",titleHighlight:"FAQ",q1:{question:"What is quantum-accelerated AI inference?",answer:"It combines classical AI inference (LLM processing) with quantum computing for specific tasks like optimization (QAOA) and random number generation (QRNG). The quantum processor handles computationally intensive sub-problems while classical GPUs handle the main inference workload."},q2:{question:"How many qubits does Infernex use?",answer:"Infernex Quantum features a 72 superconducting qubit processor. This is one of the most powerful quantum processors available via cloud API."},q3:{question:"Do I need quantum expertise to use this?",answer:"No. Our API abstracts the quantum layer. You call standard REST endpoints and our backend handles quantum circuit compilation, execution, and result processing automatically."},q4:{question:"What happens if the quantum processor is unavailable?",answer:"Automatic classical fallback. Every quantum function has a high-quality classical equivalent. Your applications never stop working."},q5:{question:"Is this real quantum computing or simulation?",answer:"Real quantum hardware. Infernex Quantum uses a physical 72-qubit superconducting quantum processor. Results come from actual quantum gate operations, not classical simulation."}},cta:{title:"The Quantum Advantage Starts at $1,999/month",subtitle:"72 qubits. QAOA + QRNG. Classical fallback included. No PhD required.",ctaQpu:"View QPU Plans",ctaVsQuant:"QPU vs Human Quant"}},td={seo:{title:"GPT API Cheaper Alternative | OpenAI Migration Guide 2026",description:"Migrate from OpenAI to Infernex in 60 seconds. Change one line of code, save 95%. Full compatibility with OpenAI SDK, LangChain, LlamaIndex. Step-by-step guide."},hero:{badge:"ð MIGRATION GUIDE",title:"Switch from OpenAI in",titleHighlight:"60 Seconds",subtitle:"Change one line of code. Save 95% on your AI inference bill. Full compatibility guaranteed.",stats:{migration:"60-second migration",costReduction:"95% cost reduction",codeChange:"1 line of code"},ctaApiKey:"Get API Key Free",ctaSavings:"Calculate Savings"},steps:{title:"3 Steps to",titleHighlight:"95% Savings",step1:{title:"Install the SDK (unchanged)",desc:"You keep the same OpenAI Python or Node.js SDK. No new dependencies."},step2:{title:"Change the Base URL",desc:"One line change. That's literally it. All endpoints work identically."},step3:{title:"Use the Same Code",desc:"Your existing code works without modification. Same models, same parameters, same response format."}},frameworks:{title:"Works with",titleHighlight:"Every Framework"},faq:{title:"FAQ",q1:{question:"Is the API really 100% compatible with OpenAI?",answer:"Yes. We implement the full OpenAI API specification: chat completions, embeddings, and more. Any code that works with OpenAI works with Infernex by changing the base URL."},q2:{question:"Which models are available?",answer:"GPT-4o, GPT-4o-mini, Claude 3.5, Llama 3.1, Mistral, Gemini, and 50+ more models â all through a single API key."},q3:{question:"How much will I save?",answer:"On average, customers save 80-95% on their AI inference costs. A typical $10,000/month OpenAI bill becomes $500-2,000/month on Infernex."},q4:{question:"Is there a free trial?",answer:"Yes. Every account starts with 1M free tokens. No credit card required."}},cta:{title:"Stop Overpaying for GPT Tokens",subtitle:"Join 500+ companies that switched from OpenAI to Infernex. Average savings: $8,000/month.",button:"Migrate Now â It's Free"}},nd={seo:{title:"How to Reduce AI Costs by 95% | Enterprise LLM Optimization",description:"Expert strategies to reduce AI costs. Our quantum-accelerated API slashes LLM inference spend by up to 95%. Drop-in OpenAI replacement. Get a cost audit."},hero:{badge:"ð° REDUCE AI COSTS BY 95%",title:"Stop Overpaying for",titleHighlight:"AI Inference",subtitle:"Same models. Same quality. 95% cheaper. Infernex is the drop-in OpenAI-compatible API that cuts your AI spending from day one.",ctaStart:"Start Free â No Credit Card",ctaSavings:"Calculate Your Savings"},savings:{title:"How Much You'll",titleHighlight:"Save",provider:"Provider",theirPrice:"Their Price",infernexPrice:"Infernex Price",youSave:"You Save"},migrate:{title:"Migrate in",titleHighlight:"2 Minutes",dropIn:{title:"Drop-in API Replacement",desc:"Change one line of code. Our API is 100% OpenAI-compatible. No SDK changes needed."},instant:{title:"Instant 95% Cost Reduction",desc:"Same models, same quality, fraction of the price. Tokens from $0.05/M."}
132,latency:{title:"Lower Latency",desc:"Sub-10ms inference latency. Faster than OpenAI, Anthropic, and Google."},sla:{title:"Enterprise SLA 99.99%",desc:"Higher uptime than any major provider. SOC 2 compliant."}},code:{title:"One Line",titleHighlight:"Change"},faq:{title:"Frequently Asked",titleHighlight:"Questions",q1:{question:"How much can I save by switching from OpenAI to Infernex?",answer:"Most customers save 90-97% on their AI inference costs. For example, GPT-4 equivalent inference costs $0.79/M tokens vs $30/M on OpenAI â a 97.4% reduction."},q2:{question:"Is the API compatible with my existing OpenAI code?",answer:"Yes, Infernex is a drop-in replacement for the OpenAI API. Change one line (the base URL) and your existing code works immediately."},q3:{question:"What about quality and accuracy?",answer:"We use the same underlying models with optimized inference infrastructure. Quality is identical â only the cost and latency are better."},q4:{question:"How long does migration take?",answer:"Under 2 minutes. Change your API base URL from api.openai.com to api.infernex.ai. That's it."},q5:{question:"Do you support enterprise-scale workloads?",answer:"Yes. We handle billions of tokens monthly with 99.99% uptime SLA, dedicated support, and custom rate limits."}},cta:{title:"Ready to Cut Your AI Costs by 95%?",subtitle:"Free tier included. No credit card required. Migrate in under 2 minutes.",button:"Start Free Now"}},ad={cover:{brand:"INFERNEX",title:"The",titleHighlight:"Cloudflare",titleEnd:"for AI Inference",subtitle:"Drop-in API replacement. Same models. 95% cheaper. <10ms latency.",seedRound:"Seed Round",location:"San Francisco · Dubai · Geneva"},problem:{tag:"The Problem",title:"AI Inference Costs Are",titleRed:"Killing",titleEnd:"Innovation",subtitle:"Companies spend 60-80% of their AI budget on inference alone. Most tokens are wasted on overqualified models.",cost:"$10-75",costDesc:"per 1M tokens (GPT-4/Claude)",waste:"80%",wasteDesc:"of requests don't need GPT-4",runway:"6 mo",runwayDesc:"avg runway lost to inference bills",quote:"Our OpenAI bill was $52K/month. At that burn rate, we had 6 months left."},solution:{tag:"The Solution",title:"Intelligent Inference",titleHighlight:"Routing",subtitle:"Infernex analyzes every prompt in real-time and routes to the optimal model â same quality, 95% less cost.",smartRouting:{title:"Smart Routing",desc:"AI analyzes complexity & routes to best model"},overhead:{title:"< 10ms Overhead",desc:"Routing adds negligible latency"},compatible:{title:"OpenAI Compatible",desc:"Change 2 lines of code. That's it."},interop:{title:"Total Interoperability",desc:"Works with any OpenAI SDK integration"}},howItWorks:{tag:"How It Works",title:"Intelligent",titleHighlight:"Model Selection",apiRequest:{label:"API Request",sub:"OpenAI-compatible"},aiClassifier:{label:"AI Classifier",sub:"Complexity analysis"},smartRouter:{label:"Smart Router",sub:"Optimal model pick"},response:{label:"Response",sub:"< 50ms total"},simpleQueries:"Simple queries",codeGen:"Code generation",complexReasoning:"Complex reasoning"},product:{tag:"Product",title:"Enterprise-Grade",titleHighlight:"Infrastructure",compatible:{title:"OpenAI SDK Compatible",desc:"Works with LangChain, LlamaIndex, any OpenAI integration"},models:{title:"50+ Models",desc:"GPT-4o, Claude, Llama, Mistral, Gemini, etc."}
132,latency:{title:"< 10ms Latency",desc:"Multi-region infrastructure, edge caching"},sla:{title:"99.99% Uptime SLA",desc:"Multi-provider redundancy, auto-failover"}},ecosystem:{tag:"Ecosystem",title:"Two",titleHighlight:"Pillars",subtitle:"Unified API for classical and quantum computing",aiInference:{badge:"AI INFERENCE",title:"95% Cheaper AI",desc:"Drop-in OpenAI replacement. From $0.05/M tokens.",feature1:"All major models",feature2:"<10ms latency",feature3:"Enterprise SLA"},quantum:{badge:"QUANTUM COMPUTING",title:"72-Qubit QPU",desc:"QAOA optimization & QRNG for finance, pharma, defense.",feature1:"72 qubits",feature2:"QAOA + QRNG",feature3:"Hybrid fallback"}},market:{tag:"Market",title:"Massive",titleHighlight:"Opportunity",tam:{value:"$150B",label:"AI Inference TAM (2027)"},cagr:{value:"47%",label:"Annual Growth"},adoption:{value:"68%",label:"Companies using AI"}},traction:{tag:"Traction",title:"Early",titleHighlight:"Momentum",users:{value:"500+",label:"Active Users"},arr:{value:"$120K",label:"ARR"},growth:{value:"35%",label:"MoM Growth"}},businessModel:{tag:"Business Model",title:"Multi-Tier",titleHighlight:"Pricing",subtitle:"Freemium to enterprise",starter:{title:"Starter",price:"Free",desc:"500K tokens/month trial"},developer:{title:"Developer",price:"$79/mo",desc:"4M tokens included"},pro:{title:"Pro",price:"$199/mo",desc:"10M tokens + priority"},enterprise:{title:"Enterprise",price:"Custom",desc:"Unlimited + SLA"}},competition:{tag:"Competition",title:"Competitive",titleHighlight:"Advantage",subtitle:"Only provider with unified classical + quantum API",openai:"OpenAI",anthropic:"Anthropic",groq:"Groq",infernex:"Infernex",costOptimization:"Cost Optimization",quantumAccess:"Quantum Access",openSource:"Open Source Friendly",no:"No",yes:"Yes",partial:"Partial"},team:{tag:"Team",title:"World-Class",titleHighlight:"Founders",ceo:{name:"Maxime Cadet",title:"CEO & Co-founder",desc:"Ex-Google AI, Stanford CS"},cto:{name:"Dr. Sarah Chen",title:"CTO & Co-founder",desc:"MIT Quantum Computing PhD"},cfo:{name:"Michael Torres",title:"CFO",desc:"Ex-Goldman Sachs, Wharton MBA"}},ask:{tag:"The Ask",title:"Raising",titleHighlight:"$3M Seed",use1:{title:"Product & Engineering",amount:"$1.5M",desc:"Expand quantum capabilities, improve routing"},use2:{title:"Go-to-Market",amount:"$1M",desc:"Enterprise sales, partnerships"},use3:{title:"Operations",amount:"$500K",desc:"Infrastructure, compliance, legal"}},contact:{tag:"Contact",title:"Let's Build the Future of",titleHighlight:"AI Inference",email:"[email protected]",website:"www.infernex.ai"},navigation:{infernex:"Infernex"}},od={features:"Features",savings:"Savings",testimonials:"Testimonials",pricing:"Pricing",faq:"FAQ",contact:"Contact",signup:"Sign Up",login:"Login",logout:"Logout",admin:"Admin",clientView:"Client View",changeTheme:"Toggle theme",models:"Models",geo:"GEO",dashboard:"Dashboard",products:"Products",openMenu:"Open menu",closeMenu:"Close menu",language:"Language",tagLaw:"Law",tagAgents:"500,000 agents",tagAiSeo:"AI SEO",tagQbit:"72 qBit",tagLightSpeed:"Light Speed",tagThermodynamics:"Thermodynamics",tagQuadFusion:"Quad Fusion",tagBenchmarks:"Benchmarks",tagConnect:"+ Connect",tagAgentDesktop:"Agent Desktop",tagCloneSite:"Clone Site",tagVoiceAi:"Voice AI",support:"Support",blog:"Blog"},rd={badge:"Ultra-Optimized AI Inference",numberOneBadge:"We are #1",titlePart1:"AI inference",titlePart2:"ultra-fast",titlePart3:"up to",titlePart4:"95% cheaper",subtitle:"Build a complete team of autonomous AI employees that work 24/7, automate complex and repetitive tasks, give your team an unprecedented level of Fortune 500-grade performance, and cut all your AI inference costs by up to 95% with intelligent LLM routing.",ctaFree:"Start Free â 50K Tokens/month",ctaSavings:"Calculate my savings",statPrice:"/ Million tokens",statPriceNote:"vs $2-8 market",statLatency:"Average latency",statLatencyNote:"Ultra low-latency",statSavings:"Savings / year",statSavingsNote:"for 1B tokens",newsBanner:"NEWS: NVIDIA Nemotron compatible now!",gatewayBadge:"4 Tools, One API",gatewayForDevelopers:"For developers",gatewayTitle:"Use 4 AI tools via one API to make your startup",gatewayTitleHighlight:"number 1 worldwide in your sector",gatewaySubtitle:"Quantum QPU, Photonics, MiroFish & TSU PBit â all through a single, unified API.",gatewayInferenceTagline:"AI Inference 95% cheaper â OpenAI-compatible API (GPT-4o, Claude, Llama, Mistral) from $0.05/M tokens.",gatewayDiscover:"Discover",pillarInference:{badge:"⦠AI INFERENCE",title:"95% Cheaper AI Inference",description:"Drop-in OpenAI-compatible API. GPT-4o, Claude, Llama, Mistral â all models, one endpoint. From $0.05/M tokens.",feature1:"95% cost reduction vs OpenAI",feature2:"<10ms latency, 99.9% uptime",feature3:"SOC 2 & GDPR, Swiss-hosted",cta:"Explore AI Inference",ctaSecondary:"Try the API"},pillarQuantum:{badge:"âï¸ QUANTUM COMPUTING",title:"72-Qubit Quantum Power",description:"QAOA optimization & QRNG for finance, pharma, and defense. The only inference API with real quantum acceleration.",feature1:"72 superconducting qubits",feature2:"QAOA portfolio optimization",feature3:"QRNG true random generation",cta:"Explore Quantum",ctaSecondary:"Quantum vs Quant"},pillarPhotonique:{badge:"ð¿ PHOTONIQUE",title:"Green Photonic AI",description:"Photonic inference (ONN): zero energy, <10µs latency. The greenest AI on the planet.",feature1:"Near-zero energy consumption",feature2:"Latency < 10 µs (speed of light)",feature3:"Zero carbon footprint",cta:"Explore Photonique",ctaSecondary:"See Plans"},pillarTsu:{badge:"TSU PBIT",title:"Thermodynamic Precision Engine",description:"Standalone TSU PBit engine for mission-critical inference: Langevin dynamics + Gibbs sampling for 99.7% factual precision.",feature1:"99.7% factual precision",feature2:"Thermodynamic p-bit sampling",feature3:"Zero-drift audit trail",cta:"Explore TSU PBit",ctaSecondary:"Precision dashboard"},pillarAllInOne:{badge:"ð¥ ALL-IN-ONE",title:"The Ultimate Bundle",description:"AI Inference + Photonique + Quantum 72 qBit. Everything in a single subscription.",feature1:"3 technologies in 1 plan",feature2:"Shared quantum brain (72 qBit)",feature3:"From $399/mo",cta:"Discover All-in-One",ctaSecondary:"Compare Plans"},pillar3in1:{badge:"INFERNEX 3-IN-1 BUNDLE",title:"3-in-1 Bundle",description:"AI Inference + Photonics + Quantum in one API. The complete stack at an unbeatable price.",feature1:"AI Inference + Photonics + Quantum",feature2:"Shared 72-qubit QPU access",feature3:"Single API key for everything",cta:"Choose 3-in-1",ctaSecondary:"See Plans"},pillarUltimate:{badge:"INFERNEX ULTIMATE SUPREMACY",title:"Ultimate Supremacy",description:"TSU PBit â All 4 pillars fused. AI + Photonics + Quantum + TSU for 99.7% precision. The Holy Grail.",feature1:"4 pillars: AI + Photonics + Quantum + TSU",feature2:"99.7% accuracy via Langevin dynamics",feature3:"Enterprise-only â corporate domains",cta:"Discover Ultimate",ctaSecondary:"TSU Calculator"}
132,pillarGeo:{badge:"GEO",title:"Infernex GEO",description:"The world's only fully automatic GEO platform. Get cited by ChatGPT, Perplexity and Gemini automatically. Track, generate, publish â in 15 languages.",feature1:"100 % automatic â zero manual work",feature2:"15 languages simultaneously",feature3:"Unlimited articles auto-published",feature4:"Real-time AI visibility score",cta:"Explore GEO",ctaSecondary:"GEO vs Competitors"},pillarMiroFish:{badge:"MIROFISH",title:"Infernex MiroFish",description:"Power your MiroFish agent swarms with the cheapest AI inference on the market. Up to 500,000 agents, $0.04/M tokens, zero config changes.",feature1:"Up to 500,000 AI agents per simulation",feature2:"95% cost reduction vs OpenAI",feature3:"< 5ms inference latency",feature4:"100% OpenAI-compatible API",cta:"Explore MiroFish",ctaSecondary:"Calculate Savings"},pillarIAO:{badge:"IAO",title:"Infernex IAO Chat",description:"The most gifted AI of its generation. Quad-Fusion architecture: MiroFish + Tsu PBit + Quantum + Photonics fused into a single, coherent response.",feature1:"4 engines fused in real-time",feature2:"2,847 agents + 72-qubit QPU",feature3:"99.7% thermodynamic precision",feature4:"Native multimodal (text, image, code, video)",cta:"Discover IAO",ctaSecondary:"Try IAO Chat"}},id={badge:"Features",title:"High-performance inference,",titleHighlight:"fraction of the cost",subtitle:"Complete optimization of your AI stack for unprecedented savings.",reduction:{title:"70-95% Reduction",description:"Massive savings on your inference costs through advanced optimization and alternative hardware."},latency:{title:"Ultra Low-Latency",description:"Less than 10ms latency for critical real-time applications."},hardware:{title:"Hardware Agnostic",description:"AMD MI300X, Intel Gaudi, NVIDIA, EPYC, edge devices. Total freedom."},quantization:{title:"Advanced Quantization",description:"4-8bit optimized models, pruning, distillation for maximum performance."},interop:{title:"Total Interoperability",description:"Compatible with all market standards. Seamless integration, no vendor lock-in."},enterprise:{title:"Enterprise Ready",description:"ISO 26262 safety, determinism, multi-GPU edge for robotics and automotive."},blackwell:{title:"Blackwell & Rubin",description:"B200 support, NVFP4, 2.8x throughput. Future-ready with Rubin (5x inference)."},benchmarks:{title:"MLPerf Benchmarks",description:"Cumulative learning on MLPerf 2026, Groq, Cerebras. Always cutting-edge."}},sd={badge:"Realized Savings",title:"Millions saved/year,",titleHighlight:"total transparency",subtitle:"Compare our rates with major cloud providers. For 1 billion tokens/year, save between",amount:"$6.8 and $7.5 million",benefits:{tokens:"Tokens from $0.05/M vs $2-8 market",migration:"No vend
132or lock-in - 5 minute migration",api:"OpenAI compatible API - instant integration",report:"Savings report generated automatically"},seeRates:"See detailed rates",tableTitle:"Cost Comparison",tableSubtitle:"Price per million tokens",provider:"Provider",theirPrice:"Their price",ourPrice:"Infernex",savingsCol:"Savings",estimateTitle:"Estimate your personalized savings",getQuote:"Get a quote"},ld={badge:"Testimonials",title:"What our clients",titleHighlight:"say about us",subtitle:"Companies of all sizes trust Infernex to reduce their AI costs",saved:"saved",perYear:"yr",stats:{clients:"Active clients",uptime:"Guaranteed uptime",latency:"Average latency",savings:"Saved/year"},items:{1:{content:"Infernex helped us reduce our AI costs by 92%. Integration was done in less than an hour thanks to the OpenAI compatible API."},2:{content:"Model quality is excellent and technical support is responsive. We migrated our entire infrastructure in one day."},3:{content:"We use Infernex for all our AI projects. The value for money is unbeatable and latency is even better than before."},4:{content:"I was skeptical at first, but the results speak for themselves. Same performance, fraction of the cost. 100% recommended."}},fallback:{one:"Infernex replaced our AI stack in one day, with much more stable latency.",two:"The bundle lets us sell several Infernex products without operational complexity."}},cd={badge:"Transparent Pricing",heroTitle:"Developer, Startup, Enterprise â choose below the ultimate fusion your SaaS needs to come alive",title:"Scale without budget explosion,",titleHighlight:"immediate savings",starter:{name:"Starter",description:"Perfect to get started and test",price:"Free",priceDetail:"50k tokens/month",features:["50k free tokens/month","Standard latency","Community support","OpenAI compatible API","1 optimized model"],note:"14-day free trial ⢠Auto-expires after 14 days",cta:"Start for free"},developer:{name:"Developer",description:"For indie devs & side projects",price:"$79",priceDetail:"/month",badge:"Best for Indies",features:["4M tokens included","5,000 API calls/month","$0.10-0.99/M extra tokens","Standard latency (~30ms)","Priority email support","All optimized models","Basic analytics"],cta:"Start with Developer"},pro:{name:"Pro",description:"For growing teams",price:"$199",priceDetail:"/month",badge:"Popular",trialBadge:"7-day free trial",guaranteeBadge:"30-day money-back guarantee",earlyAdopterBadge:"Early Adopter: +2M free tokens",earlyAdopterNote:"Bonus reserved for first 20 subscribers this month",features:["10M tokens included","$0.05-0.$49/M extra tokens","<10ms guaranteed latency","Priority support","All optimized models","Analytics dashboard","High API rate limit"],cta:"Start 7-day free trial"},scale:{name:"Scale",description:"For SaaS in growth and teams with live production traffic.",price:"$199",priceDetail:"/month",badge:"Production",features:["100M tokens included","100,000 API calls/month","Dedicated success engineer","Priority Slack support","Advanced analytics","Multi-environment API keys","99.9% SLA"],cta:"Get Scale"},business:{name:"Business",description:"For scale-ups and agencies running heavy automation.",price:"$499",priceDetail:"/month",badge:"High Volume",features:["500M tokens included","500,000 API calls/month","Custom onboarding","Dedicated Slack channel","FinOps usage reports","SSO + audit logs","99.9% SLA"],cta:"Get Business"},enterprise:{name:"Enterprise",description:"Custom solutions",price:"Custom",priceDetail:"quote",features:["Unlimited tokens","Ultra-low latency","24/7 dedicated support","99.99% SLA","⦠Semantic Cache included (<1ms)","Fine-tuning included","Dedicated infrastructure","Custom integration","Security audit"],cta:"Enterprise Request"},demoConfetti:"See successful payment animation",prepayTitle:"Pro Offer: 2 months + 1 month free",prepaySubtitle:"Save $199 by switching to quarterly billing",prepayDeadline:"Offer valid until 28.02.26",prepayButton:"Get this deal",billedInCHF:"Billed in CHF",billedInCHFTooltip:"Payment will be processed in Swiss Francs (CHF). The exact amount may vary based on exchange rates.",slotsRemaining:"Only {{count}} spots remaining!",checkoutDialog:{title:"Open Stripe payment",description:"You're in embedded preview: Stripe may be partially hidden. Open payment in a new tab.",copyLink:"Copy link",cancel:"Cancel",open:"Open"},toast:{planActivated:"ð Congratulations! Your plan has been activated!",configurationInProgress:"Configuration in progress. Contact us to subscribe.",checkoutLinkCopied:"Stripe link copied",checkoutCopyFailed:"Unable to copy automatically.",loginRequired:"Please log in to subscribe to a paid plan"},cacheAddon:{title:"Semantic Cache",badge:"Add-on · Included in Enterprise",description:"Instant responses (<1ms) on similar requests. Reduce your costs by 40% with zero config.",feature1:"<1ms cached responses",feature2:"-40% on your bill",feature3:"Works on any paid plan",perMonth:"mo",cta:"Add to plan"},ultimate:{title:"Infernex Ultimate Supremacy / TSU PBit",badge:"All-in-One",description:"Inference AI + Photonic + Quantum + TSU p-bits â the most advanced compute infrastru
132cture on Earth.",f1:"Multi-routing AI Inference",f2:"Photonic engine (light speed)",f3:"72-qubit Quantum (QAOA)",f4:"TSU p-bit sampling (99.7%)",cta:"Choose This Plan",calcBtn:"TSU Calculator"},heroSubtitle:"Choose your plan below",testimonials:{lucas:{quote:"Infernex cut our inference costs by 92%. The migration took 5 minutes.",role:"CTO, AI Startup"},rania:{quote:"We switched from OpenAI and saved $7,600/month. Same quality, fraction of the cost.",role:"VP Engineering, FinTech"},thomas:{quote:"The latency is incredible â 4ms average. Our users noticed the difference immediately.",role:"Lead Developer, SaaS Platform"}},standaloneTitle:"Autonomous Products",payg:{name:"Pay-as-you-go",description:"No monthly fee. Pay only for what you use â $0.20 per million tokens.",price:"$0.20",priceDetail:"/ 1M tokens",cta:"Start pay-as-you-go",f1:"No monthly commitment â cancel anytime",f2:"60% cheaper than OpenAI gpt-4o-mini blended",f3:"Configurable hard cap (default $100/mo) â no surprises",f4:"Same models, streaming enabled by default",f5:"Consolidated monthly invoice"}},dd={seo:{title:"API Playground - Infernex",description:"Test the Infernex API directly in your browser. <10ms latency, unbeatable prices."},badge:"Playground",title:"Test the API",subtitle:"Experiment with our API directly in your browser",request:{title:"Request",description:"Select a model and enter your prompt"},response:{title:"Response",description:"Result of your API request"},labels:{model:"Model",prompt:"Prompt"},placeholders:{prompt:"Enter your prompt here..."},examples:{label:"Examples:"},examplePrompts:{summary:{label:"Summary",prompt:"Summarize the benefits of AI inference optimization in 3 points."},code:{label:"Code",prompt:"Write a Python function to calculate savings between two API rates."},analysis:{label:"Analysis",prompt:"Compare the pros and cons of different LLM model architectures."}},actions:{running:"Running...",run:"Run"},metrics:{latency:"Latency",tokens:"Tokens",cost:"Cost"},output:{generating:"Generating...",placeholder:"Response will appear here..."},cta:{title:"Ready to integrate?",subtitle:"Get your API key and start saving today.",button:"Get my API key"},toasts:{enterPrompt:"Please enter a prompt",apiError:"API call error",codeCopied:"Code copied!"},demoResponse:{intro:"Here's a demo response for the {{model}} model.",body:"This interface allows you to test the Infernex API directly in your browser.",production:"In production, you would see the real model response with ultra-low latency (<10ms).",howToUse:"To use the API in production:"}},ud={seo:{title:"Price Comparator - Infernex",description:"Compare your current AI costs with Infernex. Up to 95% reduction on inference costs."},badge:"Comparator",titlePrefix:"Compare your ",titleHighlight:"costs",subtitle:"Compare your current costs with Infernex and discover how much you can save",input:{title:"Your current usage",description:"Enter your usage parameters"},labels:{currentProvider:"Current provider",model:"Model",tokensPerMonth:"Tokens per month",orEnterManually:"Or enter manually"},tokensUnit:"M tokens",results:{title:"Your savings"},perMonth:"/month",annualSavings:"Annual savings",breakdown:{monthlySavings:"Monthly savings",infernexPrice:"Infernex price",providerPrice:"{{provider}} price",addOns:"Add-ons"},actions:{getStarted:"Get started"},addons:{bundleDesc:"3-in-1 Bundle: Inference + Quantum + Photonique",bundleHint:"Save 40% compared to individual products",photoniqueDesc:"Photonic co-processing for ultra-fast inference",quantumDesc:"Quantum-enhanced optimization layer",subtitle:"Enhance your plan with premium add-ons",title:"Add-ons"}},md={badge:"FAQ",title:"Frequently",titleHighlight:"asked questions",subtitle:"Everything you need to know about our optimized AI inference service.",q1:{question:"What is optimized AI inference?",answer:"Optimized AI inference means running artificial intelligence models more efficiently, reducing costs and latency while maintaining result quality. Our technology uses advanced quantization and optimization techniques to achieve up to 95% savings."},q2:{question:"How does the $0.05-0.79/M token pricing work?",answer:"Our pricing is based on the number of tokens processed. A token represents about 4 characters. Depending on the model used, the price ranges from $0.05 to $0.79 per million tokens, compared to $2 to $8 with traditional providers. You only pay for what you use."},q3:{question:"How does the 2 months + 1 free offer work?",answer:"When subscribing to the quarterly Pro plan, you pay for 2 months and the 3rd month is free. That's $199 in direct savings. Simply click the 'Get this deal' button in the Pricing section to subscribe directly online."},q4:{question:"What is the average latency of your APIs?",answer:"Our infrastru
132cture guarantees an average latency of less than 10ms for most requests. We use optimized GPU servers and distributed architecture to ensure ultra-fast response times."},q5:{question:"Is the API compatible with OpenAI?",answer:"Yes, our API is 100% compatible with OpenAI format. You can migrate your existing applications by simply changing the endpoint URL. No code modification required."},q6:{question:"Do you offer a free trial?",answer:"Yes, our Starter plan includes 50,000 free tokens for 14 days, no commitment required. It's ideal for testing our service and evaluating potential savings before upgrading to a paid plan."},q7:{question:"How are annual savings calculated?",answer:"Savings are calculated by comparing the cost of your tokens with us versus standard market rates (OpenAI, Anthropic, etc.). For 1 billion tokens per year, savings can reach $6.8M depending on the models used."},q8:{question:"What AI models are available?",answer:"We offer a wide range of optimized models including alternatives to GPT-4, Claude, and Llama. All our models are fine-tuned to offer the best value for money. The Enterprise plan also allows custom fine-tuning."},q9:{question:"What does the 30-day money-back guarantee cover?",answer:"If our service doesn't reduce your AI inference costs compared to your previous provider within 30 days, we refund you in full â no questions asked. Simply provide proof of your previous costs, and we handle the rest. This applies to all paid plans."},q10:{question:"What is your SLA and uptime guarantee?",answer:"We guarantee 99.9% uptime backed by our multi-provider failover architecture. If one provider goes down, your requests are automatically routed to another in under 200ms. This gives you higher reliability than any single AI provider can offer, including OpenAI directly."},q11:{question:"How do you handle data security and GDPR compliance?",answer:"Your data is encrypted in transit (TLS 1.3) and at rest. We never store your prompts or responses â they are processed in real-time and discarded. Our infrastructure is GDPR-compliant with data processing in European data centers. No training is ever done on your data."},q12:{question:"How do I use a promo code?",answer:"Enter your promo code during checkout to instantly apply the discount. Active promotions are displayed in the banner at the top of the page. Promo codes can offer percentage discounts, free months, or bonus tokens depending on the current offer."},q13:{question:"Can I test Infernex without a credit card?",answer:"Yes. Sign up for the Starter plan and get 50,000 free tokens per month â no card required. If you need more, join the Founders Club at â¬59/month for life (500K tokens/month) or switch to pay-as-you-go at $0.20 per 1M tokens with no minimum."}},pd={badge:"Contact",title:"Let's talk about your",titleHighlight:"AI savings",subtitle:"Our team is available to help you optimize your inference infrastructure.",general:{title:"General Questions & Partnerships",description:"Quotes, demonstrations, integrations, business proposals."},support:{title:"Technical Support",description:"API issues, bugs, implementation questions, debugging."},madeWith:"Made with"},fd={newsletter:{title:"Stay at the forefront of AI inference",subtitle:"Receive our news, benchmarks and optimization tips directly in your inbox.",placeholder:"[email protected]",subscribe:"Subscribe",subscribed:"Subscribed!",noSpam:"No spam. Unsubscribe in one click."},tagline:"The future of inference is fast, efficient, and affordable.",product:"Product",resources:"Resources",compare:"Compare",playground:"Playground",calculator:"Comparator",connect:"Connect",status:"Status",changelog:"Changelog",company:"Company",about:"About",blog:"Blog",careers:"Careers",press:"Press",legal:{privacy:"Privacy",terms:"Terms",cookies:"Cookies"},documentation:"Documentation",allRights:"All rights reserved.",vsOpenAI:"vs OpenAI",vsGroq:"vs Groq",vsTogether:"vs Together AI",vsNvidia:"vs NVIDIA",vsAnyscale:"vs Anyscale",partners:"Partners",investors:"Investors",testimonials:"Testimonials",support:"Support"},hd={seo:{title:"AI Inference Pricing 2026 â Compare All LLM Costs | Infernex",description:"AI inference pricing comparison. GPT-4, Claude, Llama, Mistral, Gemini rates side-by-side. Discover how to reduce costs by 95% with Infernex's optimized inference."},hero:{badge:"ð AI INFERENCE PRICING",title:"The Most",titleHighlight:"Trans
132parent Pricing",titleEnd:"in AI",subtitle:"Compare all major LLM providers side-by-side. See exactly how much you'll save with Infernex's quantum-accelerated inference.",ctaCalculator:"Calculate My Savings",ctaFree:"Start Free",title_p1:"AI Inference:",title_p2:"Pricing",title_p3:"Compared"},comparison:{title:"Provider",titleHighlight:"Comparison",subtitle:"All prices per 1 million tokens (input + output average)",provider:"Provider",model:"Model",pricePerM:"Price / 1M tokens",infernexPrice:"Infernex Equivalent",savings:"You Save"},why:{title:"Why",titleHighlight:"Infernex Costs Less",routing:{title:"Intelligent Routing",desc:"80% of requests don't need GPT-4. Our AI router picks the optimal model for each request, slashing unnecessary costs."},hardware:{title:"Alternative Hardware",desc:"AMD MI300X, Intel Gaudi3, edge TPUs â we use the most cost-effective hardware for each workload instead of expensive NVIDIA-only setups."},quantum:{title:"Quantum Optimization",desc:"72-qubit Infernex Quantum processor accelerates model selection, caching decisions, and batch scheduling â cutting compute waste."},noMiddleman:{title:"No Middleman Markup",desc:"We negotiate direct volume deals with model providers and pass the savings to you. No Azure/AWS markup layers."}},calculator:{title:"Live",titleHighlight:"Savings Calculator",currentSpending:"Your current monthly AI spend",provider:"Current provider",estimatedSavings:"Estimated monthly savings",perYear:"per year",cta:"Get Started Free"},faq:{title:"Pricing",titleHighlight:"FAQ",q1:{question:"How can Infernex be 95% cheaper than OpenAI?",answer:"We don't use a single expensive provider. Instead, we intelligently route requests across 50+ optimized models. A simple query that OpenAI charges $30/1M tokens for can run on a Llama model at $0.10/1M with identical quality."},q2:{question:"Is there a quality trade-off?",answer:"No. Our router ensures quality parity. If a cheaper model can't handle your request with 95%+ accuracy, we automatically escalate to a premium model. You get the same results, just at a fraction of the cost."},q3:{question:"What if I have spiky usage?",answer:"Perfect for spiky workloads. You only pay per token â no idle costs. Process 100M tokens one month and 10M the next. You're never locked into expensive minimums like AWS Bedrock."},q4:{question:"Can I lock in these prices long-term?",answer:"Yes. Enterprise plans offer 12-36 month price locks. We guarantee your per-token rate won't increase even if upstream providers raise prices."},q5:{question:"Do you offer volume discounts?",answer:"Yes. Enterprise plans start at custom pricing for 1B+ tokens/month. Contact us for a tailored quote with volume commitments and SLA guarantees."},a1:"AI inference pricing is the cost charged per token (or per query) when using a model via API. Providers like OpenAI charge between $5 and $30 per million tokens, while Infernex offers the same models starting at $0.05/M.",a2:"We use an optimized infrastructure with semantic caching, model quantization, and efficient GPU scheduling."},cta:{title:"Ready to Cut Your AI Bill by 95%?",subtitle:"Free tier included. No credit card required. Migrate in under 2 minutes.",button:"Start Saving Today"},button:{calculate:"Calculate your savings",try_free:"Free trial"}},gd={login:{title:"Login",subtitle:"Access your Infernex account",email:"Email",password:"Password",submit:"Sign in",loading:"Signing in...",noAccount:"Don't have an account?",createAccount:"Create an account",continueGoogle:"Continue with Google",or:"or",success:"Login successful!",googleError:"Error signing in with Google"},signup:{title:"Create an account",subtitle:"Sign up to access Infernex",email:"Email",password:"Password",confirmPassword:"Confirm password",submit:"Create my account",loading:"Creating...",hasAccount:"Already have an account?",signIn:"Sign in",continueGoogle:"Continue with Google",or:"or",success:"Account created successfully!",passwordMismatch:"Passwords don't match",passwordTooShort:"Password must be at least 6 characters",passwordCompromised:"This password has been found in data breaches. Please choose a different password for your security.",googleError:"Error signing in with Google"},verify:{title:"Verify your email",subtitle:"A 6-digit code has been sent to",submit:"Verify my email",loading:"Verifying...",noCode:"Didn't receive the code?",resend:"Resend",back:"Back to signup",codeSent:"A verification code has been sent to your email",invalidCode:"Invalid or expired code",success:"Email verified successfully!",resendSuccess:"A new code has been sent",resendError:"Error resending code"}},bd={title:"Agent Dashboard",welcome:"Welcome,",loading:"Loading...",tabs:{overview:"Overview",api:"API Integration",savings:"Savings",referral:"Referral"},plan:{title:"My Subscription",subtitle:"Your plan details and usage",free:"Free",nextBilling:"Next billing",tokensUsed:"Tokens used this month",used:"used",remaining:"remaining",upgrade:"Upgrade to",manage:"Manage my subscription"},stats:{title:"Quick Statistics",subtitle:"Your usage metrics this month",tokensUsed:"Tokens used",apiCalls:"API calls",avgLatency:"Average latency",savingsVsOpenAI:"Savings vs OpenAI"},api:{title:"API Integration",subtitle:"OpenAI SDK compatible ⢠Just change the endpoint",apiKey:"API Key",endpoint:"Endpoint",codeExamples:"Code examples",copied:"copied!",planBadge:"Plan"},quickstart:{title:"Quick Start",subtitle:"Integrate Infernex in 5 minutes ⢠OpenAI SDK compatible",step1:{title:"Get your API key",description:"Copy your API key from the section above"},step2:{title:"Install the SDK",description:"pip install openai or npm install openai"},step3:{title:"Change the endpoint",description:"Point to api.infernex.ai instead of OpenAI"},step4:{title:"Test your first request",description:"Send a request and verify the response"},includes:"includes:",viewDocs:"View full documentation"},savingsCalc:{title:"Realized Savings",subtitle:"Real-time comparison with your",tokens:"tokens",yourCost:"Your Infernex cost",totalSaved:"Total saved vs average"},referral:{title:"Referral Program",subtitle:"Earn 20% commission on each referral",code:"Your referral code",shareLink:"Share this link:",referrals:"Signed up referrals",earnings:"Total earnings",affiliateBadge:"Verified affiliate",commission:"commission",howItWorks:"How it works?",step1:"Share your unique link with your contacts",step2:"They sign up and subscribe to a plan",step3:"You earn 20% of their subscription, every month!",linkCopied:"Referral link copied!"},alerts:{suspended:"Access suspended â Quota exceeded",suspendedMsg:"Your monthly token quota is exhausted. To continue using our services without interruption, please upgrade to a higher plan. Your API access will be restored once you upgrade.",critical:"Token limit almost reached",criticalMsg:"You have used {{percent}}% of your tokens. Upgrade to a higher plan to avoid interruptions.",warning:"High token usage",warningMsg:"You have used {{percent}}% of your monthly token quota.",apiSuspended:"Access suspended â Call limit reached",apiSuspendedMsg:"You have reached the maximum number of API calls allowed this month. To restore access to our services, please subscribe to a plan that suits your needs.",apiCritical:"API call limit almost reached",apiCriticalMsg:"You have made {{percent}}% of your monthly API calls.",apiWarning:"High API call count",apiWarningMsg:"You have used {{percent}}% of your API call quota.",upgradeNow:"Upgrade to Pro now",upgradePro:"Upgrade to Pro",contactEnterprise:"Contact for Enterprise",upgradeEnterprise:"Upgrade to Enterprise",close:"Close"},noBillingDate:"No billing date set",connecting:"Not connected to the local agent",connected:"Connected",disconnected:"Disconnected",reconnect:"Reconnect",reconnecting:"Reconnecting...",reconnect_failed:"Reconnection failed",token_placeholder:"Paste your token here",connect_btn:"Connect",local_only:"Local Only",quad_fusion_active:"Quad Fusion Active",nav:{status:"Status",chat:"Chat",mail:"Mail",terminal:"Terminal",files:"Files",calendar:"Calendar",notifications:"Notifications",skills:"Skills",logs:"Logs",multiagent:"Multi-Agent",browser:"Browser",smarthome:"Smart Home",security:"Security",settings:"Settings"},voice:{listening:"Listening...",speak_now:"Speak now",processing:"Processing...",not_supported:"Voice not supported on this browser",auto_submit:"Auto-submit voice input",read_responses:"Read responses aloud",stop_reading:"Stop reading",hold_to_talk:"Hold space to talk"},chat:{placeholder:"Ask your agent anything...",send:"Send",new_session:"New Session",thinking:"Thinking...",tool_running:"Running tool...",route_to_quad:"Route to Quad Fusion"},mail:{read_inbox:"Ask your agent to read your inbox",summarize:"Summarize",compose:"Compose",reply:"Reply",forward:"Forward",archive:"Archive",delete:"Delete"},terminal:{placeholder:"Type a command or ask in natural language...",ask_agent:"Ask Agent",stop:"Stop",warning_destructive:"This command may be destructive",confirm_execute:"Confirm Execution"},files:{open:"Open",rename:"Rename",move:"Move",delete:"Delete",create:"Create",analyze_quad:"Analyze with Quad Fusion"},approval:{title:"Action Approval Required",approve:"Approve",deny:"Deny",action_label:"Pending Action"}},yd={title:"Admin Dashboard",welcome:"Welcome, {{email}}. Complete view of your activity (31 manifesto points).",loading:"Loading...",logout:"Logout",clientView:"Client View",tabs:{overview:"Overview",enterprise:"Enterprise",growth:"Growth",finance:"Finance",gaming:"Gaming",innovation:"Innovation"},clients:{title:"Client Management",addClient:"Add Client"},goals:{title:"Monthly Goals",profitMin:"Profit Min",newClients:"New Clients",adsBudget:"Ads Budget",overallProgress:"Overall goal achieved",noGoals:"No goals defined for this month"},gaming:{moduleTitle:"Gaming Module - Point 24",features:{activation:"Automatic activation at 15 paying clients",latency:"Latency optimizations < 5ms guaranteed",plugins:"Auto-generated Unity/Unreal/Godot plugins",landing:"Dedicated gaming landing page",acquisition:"Targeted game dev acquisition"}},innovation:{moduleTitle:"Cumulative Apple Innovation - Point 25",features:{micro:"Micro-innovations: 1-3 weeks",major:"Major innovations: 1-4 months",announce:"Landing banner announcement 7-14 days before",tracking:"Revenue tracking per innovation",alert:"Alert if > 21 days without innovation"},statusTitle:"Points 26-31 Status",statuses:{microsoft:"Microsoft 365/Copilot Disruption - Active",multilingual:"Multilingual 50+ languages - Active",gafa:"GAFA acquisition tactics - Active",freemium:"Freemium/GDPR - /privacy page live",layers:"Layers Extension - Infrastru
132cture ready"}}},vd={title:"Enterprise Request",dialogTitle:"Enterprise Plan Request",dialogDescription:"Get a personalized quote tailored to your needs in less than 24h.",companyNameRequired:"Company name required",contactNameRequired:"Contact name required",invalidEmail:"Invalid email",volumeRequired:"Estimated volume required",labels:{company:"Company *",companyPlaceholder:"Your company name",contact:"Contact name *",contactPlaceholder:"Your name",email:"Professional email *",emailPlaceholder:"[email protected]",phone:"Phone",volume:"Estimated monthly volume *",volumePlaceholder:"Select volume",useCase:"Use case",useCasePlaceholder:"Project type",specificNeeds:"Specific needs",specificNeedsPlaceholder:"Describe your needs: integration, SLA, dedicated infrastructure, fine-tuning..."},volumes:{"50m":"50M tokens/month","100m":"100M tokens/month","500m":"500M tokens/month","1b":"1B+ tokens/month"},useCases:{chatbot:"Chatbot / AI Assistant",gaming:"Gaming / Real-time NPC",automation:"Workflow automation",analytics:"Analytics / BI",content:"Content generation",other:"Other"},benefits:{latency:"Latency <5ms",latencyDesc:"Dedicated infrastructure",sla:"SLA 99.99%",slaDesc:"Contractual guarantee",team:"Dedicated team",teamDesc:"Assigned engineer",support:"24/7 Support",supportDesc:"Maximum priority"},priceEstimate:"Price estimate",startingFrom:"Starting from",perMonth:"/month",savings:"savings",priceNegotiable:"Final price negotiable based on your specific needs",submit:"Submit request",submitting:"Sending...",bookDemo:"Book a demo",successTitle:"Request sent!",successMessage:"Our team will c
132ontact you within 24h with a personalized quote.",successToast:"Request sent! We will contact you within 24h.",errorToast:"Error sending request. Please try again.",disclaimer:"By submitting this form, you agree to be contacted by our sales team."},Id={title:"We use cookies",description:'We use cookies to improve your experience, analyze traffic and personalize content. By clicking "Accept all", you consent to the use of all cookies.',privacyLink:"Privacy Policy",acceptAll:"Accept all",necessaryOnly:"Necessary only",customize:"Customize",close:"Close",preferencesTitle:"Cookie preferences",necessary:{title:"Necessary cookies",description:"Essential for site functionality (authentication, security)",badge:"Always active"},analytics:{title:"Analytics cookies",description:"Help us understand how you use our site"},marketing:{title:"Marketing cookies",description:"Allow personalized ads"},savePreferences:"Save my preferences"},Ad={badge:"ROI CALCULATOR",title:"Calculate Your",titleHighlight:"Real Savings",subtitle:"See exactly how much you'll save by switching from your current provider to Infernex.",providerLabel:"Current Provider",tokensLabel:"Monthly Token Volume (M)",tokensUnit:"M = Million tokens",generate:"Calculate ROI",annualSavings:"Annual Savings",projection3Y:"3-Year Projection",savedOver3Y:"saved over 3 years",downloadPDF:"Download PDF Report",startSaving:"Start Saving Now",pdfTitle:"ROI Report - Infernex vs Your Provider",pdfGenerated:"Generated on:",currentProvider:"Current Provider Cost",cheaper:"cheaper",savings:"Savings",monthly:"Monthly",annual:"Annual",threeYear:"3-Year",totalSavings3Y:"Total 3-Year Savings",pdfFooter:"The 95% cheaper AI inference API"},xd={badge:"â¡ ENTERPRISE SLA",title:"99.99%",titleHighlight:"Uptime Guarantee",subtitle:"Higher reliability than any major AI provider. Automated failover across multiple providers ensures your apps never go down.",uptimeLabel:"Guaranteed uptime",downtimeMonth:"downtime/month max",failoverTime:"auto-failover",monitoring:"monitoring",compensationTitle:"Service Credits",compensationDesc:"Automatic compensation if we miss our SLA targets:",uptimeBelow:"Uptime drops below",creditLabel:"Service Credit",autoApplied:"Credits applied automatically to your next invoice"},wd={badge:"Industries We Serve",title:"Powering AI Across Industries",subtitle:"From SaaS to FinTech, we help companies cut their LLM costs by up to 95%",saas:{label:"SaaS",desc:"AI-powered features, chatbots, content generation"},fintech:{label:"FinTech",desc:"Fraud detection, risk analysis, customer support"},devtools:{label:"DevTools",desc:"Code completion, documentation, automated testing"},gaming:{label:"Gaming",desc:"NPC dialogue, dynamic storytelling, player support"},robotics:{label:"Robotics",desc:"Real-time decision making, sensor processing"},autonomous:{label:"Autonomous Vehicles",desc:"Object detection, path planning, safety systems"},cta:{text:"Don't see your industry? We serve any business using LLM APIs.",button:"Get Custom Quote"}},Sd={title:"30-Day Money-Back Guarantee",description:"Not satisfied? We'll refund you, no questions asked. We're confident you'll save 80-95% on your AI costs.",conditions:["Full refund within 30 days","No questions asked","Available on all paid plans","Instant processing"],cta:"Start Risk-Free Trial"},kd={explainer:{title:"Use Cases & Proven Enterprise Gains",intro:"Infernex Quantum is the world's first commercial 72-qubit inference brain, delivered via API. It solves optimization, routing and simulation problems that classical GPUs cannot approach â for a fraction of the cost of IBM or Google Quantum.",usageTitle:"Concrete Business Use Cases",usages:["Finance: portfolio optimization on 5,000+ assets in <1s (vs hours on GPUs).","Logistics: real-time route optimization for 10,000+ deliveries daily.","Pharma & materials: molecular simulation at 99.2% precision for drug discovery.","Cybersecurity: post-quantum key testing and lattice cryptanalysis at scale."],gainsTitle:"Measured Enterprise Gains",gains:["99.2% precision on QAOA benchmarks â matching Google Quantum AI at 1/200th the price.","530k+ verified quantum operations already served across 145 countries.","â94% cost vs a $10M+ on-prem quantum lab â pay-as-you-go from $27,440/mo.","Sub-second QPU allocation via Origin Pilot Wukong v2 â no queue, no waitlist."],revolution:"72 real qubits, not simulation. This is not classical AI accelerated â it is a physical shift in what is computable. Problems that took 6 weeks on H100 clusters now finish in 4 seconds.",superiority:"Why Infernex wins: we own the only commercial quantum API that pairs 72 qubits with Photonic and MiroFish fallbacks. IBM sells hardware. Google sells research. We sell working results."},sectors:{finance:"Finance & Banking",family_office:"Family Office",private_banking:"Private Banking",pharma:"Pharma & Biotech",defense:"Defense & Security",research:"Scientific Re
132search",energy:"Energy & Utilities",crypto:"Crypto & DeFi",hft:"High-Frequency Trading",insurance:"Insurance"},seo:{title:"Infernex Quantum | QAOA & QRNG AI Optimization â 72 Qubits",description:"Access high-performance quantum computing. QAOA and QRNG with 72 qubits for enterprise AI. Inference 95% cheaper through quantum optimization.",keywords:"quantum computing, Infernex Quantum, QAOA, QRNG, 72 qubits, quantum inference, quantum computing finance, accessible quantum computing",ogTitle:"Infernex Quantum â Accessible Quantum AI",ogDescription:"QAOA, QRNG, 72 qubits. Quantum computing for finance, pharma, and defense. From $1,999/month."},hero:{badge:"⦠QUANTUM â WORLD FIRST",title:"Quantum AI",titleHighlight:"Accessible with Infernex Quantum and 99.2% precision.",description:"Access the power of quantum computing through Infernex Quantum â 72 qubits, QAOA optimization, and QRNG â all integrated into our inference infrastructure.",sectors:["Banking","Family Office","Pharma","Defense","Crypto","HFT","Insurance"],stats:{qubits:"Qubits",precision:"Precision",countries:"Countries",operations:"Operations"},ctaPlans:"View Plans",ctaROI:"Calculate My ROI",imgAlt:"Quantitative analyst using Infernex Quantum QPU for financial optimization"},stats:{quantumOps:"Quantum operations",countries:"Countries covered",visits:"Visits/month",qubitsAvailable:"Qubits available"},manifesto:{badge:"⦠MANIFESTO",title:"If you have to stop at the price,",titleHighlight:"this product is not for you.",titleSub:"And you don't understand the stakes of the quantum world at your doorstep to be the leader in your sector.",description:"Google Willow Quantum: not available to the public. IBM Quantum Platform (Qiskit): minimum $40,000/month. Infernex QPU Enterprise: $27,440/month â the most affordable on the market.",difference:"The difference? We democratize quantum computing.",codeComment:"# 3 lines. That's it.",roi:"ROI $1B+",roiDesc:"Cumulative value generated by clients and Enterprise using quantum technology",willowStatus:"not available to the public",ibmStatus:"minimum $40,000/month",cheapest:"the most affordable on the market"},vision:{badge:"⦠QUANTUM VISION",title:"We democratized quantum computing",titleHighlight:"for you.",p1:"We offer you a solution with reliable scenarios proven at 99.2% precision for all your analyses, decisions, and solutions.",p2:"You become #1 in your sector and dominate your competitors.",p3:"It's no longer a matter of choice â it's a matter of interest and leadership.",question:"Do you want to become â or remain â the leader in your sector?",answer:"If so, Infernex Quantum becomes your quantum brain as the ultimate analyst, 24/7, always active.",cta:"Activate Your Quantum Brain",firstDate:"And we were the first to do it. 03/01/2026",step1:"Sign up",step2:"Pay for your plan",step3:"Receive your API Keys"},personas:{title:"Who is",titleHighlight:"Infernex Quantum",titleSuffix:"for?",subtitle:"11 decision-maker profiles transforming their industry with quantum computing.",items:{ctoFinance:{title:"CTO Finance",desc:"Optimize your risk models with quantum computing"},cioFamily:{title:"CIO Family Office",desc:"QAOA asset allocation for superior returns"},headQuant:{title:"Head Quant",desc:"Replace 10 quants with a single QPU"},rdPharma:{title:"R&D Dir. Pharma",desc:"Molecular simulation 1000x faster"},cisoDef:{title:"CISO Defense",desc:"Post-quantum cryptography and QRNG"},labDir:{title:"Lab Director",desc:"Combinatorial optimization for research"},ctoEnergy:{title:"CTO Energy",desc:"Real-time energy grid optimization"},headCrypto:{title:"Head Quant Crypto",desc:"High-frequency DeFi quantum trading"},ctoHft:{title:"CTO HFT",desc:"Sub-microsecond latency with QRNG"},caoIns:{title:"CAO Insurance",desc:"Quantum actuarial modeling"},futureQuant:{title:"Future Quant Hire",desc:"Why recruit when you can automate?"}},learnMore:"Learn More â"},whyQuantum:{title:"Why",titleHighlight:"Quantum",titleSuffix:"?",sectors:{finance:{title:"Finance",desc:"QAOA portfolio optimization, quantum risk models, accelerated Monte Carlo options pricing."},pharma:{title:"Pharma",desc:"Molecular simulation, drug discovery 1000x faster, protein folding."},defense:{title:"Defense",desc:"Post-quantum cryptography, QRNG anomaly detection, logistics optimization."},research:{title:"Research",desc:"Combinatorial optimization problems, complex system simulation, quantum machine learning."}
132,crypto:{title:"Crypto & HFT",desc:"Quantum algorithmic trading, cross-chain QAOA arbitrage, verifiable random number generator."},insurance:{title:"Insurance",desc:"Actuarial modeling, quantum fraud detection, reserve optimization."}}},howItWorks:{title:"How it",titleHighlight:"works",titleSuffix:"?",steps:{s1:{title:"Send your request",desc:"Via our REST API or Python SDK. Specify the operation type: QAOA, QRNG, or simulation."},s2:{title:"Intelligent routing",desc:"Our infrastructure automatically routes to the Infernex Quantum processor (72Q) or an optimized classical fallback."},s3:{title:"Real-time results",desc:"Receive optimized results with metadata: quantum vs classical, execution time, confidence."}},compareLink:"Compare with a Quant Associate â"},planDiff:{title:"Pro vs",titleHighlight:"Enterprise",aiModels:{title:"AI Models",pro:"GPT-4, Llama, Mistral",enterprise:"All + Dedicated QPU"},quantumCompute:{title:"Quantum Computing",pro:"Shared QAOA",enterprise:"Dedicated 72Q QPU"},support:{title:"Support",pro:"Email 24h",enterprise:"Dedicated Slack + 99.99% SLA"}},roiCalc:{title:"Quantum",titleHighlight:"ROI Calculator",subtitle:"Estimate the return on investment of quantum AI for your sector.",sectorLabel:"Sector",assetsLabel:"Assets under management / R&D Budget ($)",gainLabel:"Estimated gain/year",costLabel:"Annual cost",roiLabel:"ROI"},apiCode:{title:"Integration",titleHighlight:"in 3 lines",comment:"# Python SDK"},transparency:{title:"Total",titleHighlight:"transparency",responsibilities:{title:"Our responsibilities",items:["⦠Guaranteed uptime 99.9% (99.99% Enterprise)","⦠Automatic classical fallback if QPU unavailable","⦠AES-256 encrypted data in transit and at rest","⦠No data retention post-processing"]},commitments:{title:"Our commitments",items:["⦠14-day money-back guarantee","⦠No lock-in â export your data anytime","⦠Transparent pricing with no hidden fees","⦠Human support, no chatbot"]},disclaimer:"Disclaimer: Quantum performance depends on the availability of the Infernex Quantum processor and the nature of the submitted problem. In case of unavailability, an optimized classical fallback is automatically activated. ROI projections are estimates based on market data and do not constitute financial advice."},plans:{title:"Quantum",titleHighlight:"Plans",monthly:"Monthly",annual:"Annual",annualDiscount:"-20%",pro:{name:"Quantum Pro",perMonth:"/month",guarantee:"14-day money-back guarantee",cta:"Subscribe Quantum Pro"},enterprise:{name:"QPU Enterprise",recommended:"Recommended",perMonth:"/month",subtitle:"Average ROI $1B+ · 99.2% precision · 0 competitors",cta:"Subscribe QPU Enterprise"},proFeatures:["Shared QAOA access","Unlimited QRNG","Python SDK + REST API","72 qubits","Semantic cache","Email support 24h","Real-time monitoring","JSON/CSV exports","10M tokens/month","Classical fallback","Multi-region","Full documentation"],proExcluded:["Dedicated QPU","Dedicated Slack","99.99% SLA","Unlimited tokens","Personalized onboarding"],enterpriseFeatures:["Dedicated 72-qubit QPU","Priority QAOA","Unlimited QRNG","Python + REST + WebSocket SDK","Unlimited tokens","99.99% SLA","Dedicated Slack + Account Manager","Personalized onboarding","Priority multi-region","Security audit included","Custom exports","High API rate limit","Advanced monitoring","Team training","Quarterly review"],disclaimer:{title:"{{plan}} Terms",body1:"By subscribing to the {{plan}} plan, you accept that quantum performance depends on the availability of the Infernex Quantum processor. An optimized classical fallback is automatically activated in case of unavailability.",body2:"ROI projections are estimates and do not constitute financial advice. Full refund within the first 14 days if not satisfied.",acceptLabel:"I have read and accept the terms",cancel:"Cancel",loading:"Loading...",proceed:"Proceed to payment"},loginRequired:"Log in to subscribe"},comparison:{title:"Comparison",titleHighlight:"table",feature:"Feature",pro:"Pro",enterprise:"Enterprise",rows:{qubits:{feature:"Available qubits",pro:"72 (shared)",enterprise:"72+ (dedicated)"},qaoa:{feature:"QAOA",pro:"â",enterprise:"â Priority"},qrng:{feature:"QRNG",pro:"â",enterprise:"â"},tokens:{feature:"Tokens/month",pro:"10M",enterprise:"Unlimited"},latency:{feature:"Average latency",pro:"<200ms",enterprise:"<50ms"},sla:{feature:"SLA Uptime",pro:"99.9%",enterprise:"99.99%"},support:{feature:"Support",pro:"Email",enterprise:"Dedicated Slack"},am:{feature:"Account Manager",pro:"â",enterprise:"â"},onboarding:{feature:"Onboarding",pro:"Self-service",enterprise:"Personalized"},audit:{feature:"Security audit",pro:"â",enterprise:"Included"},multiRegion:{feature:"Multi-region",pro:"â",enterprise:"Priority"},sdk:{feature:"SDK",pro:"Python + REST",enterprise:"All + WebSocket"},cache:{feature:"Semantic cache",pro:"â",enterprise:"â Advanced"},fallback:{feature:"Classical fallback",pro:"â",enterprise:"â"},training:{feature:"Training",pro:"Docs",enterprise:"On-site"},review:{feature:"Quarterly review",pro:"â",enterprise:"â"},price:{feature:"Price/month",pro:"$1,999",enterprise:"$27,440"}}},faq:{title:"Frequently Asked",titleHighlight:"Questions",items:{q1:{question:"What is Infernex Quantum?",answer:"Infernex Quantum is a quantum computing platform accessible via REST API. It integrates a 72-qubit superconducting processor for QAOA optimization and QRNG random number generation, with automatic classical fallback."},q2:{question:"What is QAOA and how does it work?",answer:"QAOA (Quantum Approximate Optimization Algorithm) solves combinatorial optimization problems â portfolio allocation, logistics routing, scheduling â with precision impossible to achieve classically. It runs on our 72-qubit processor."},q3:{question:"What is QRNG used for?",answer:"QRNG (Quantum Random Number Generator) produces truly random numbers using quantum mechanics. Applications: unbreakable encryption keys, unbiased A/B testing, quantum Monte Carlo for options pricing, and seeds for DeFi smart contracts."},q4:{question:"Do I need quantum physics expertise?",answer:"No. Our API fully abstracts the quantum layer. You call standard REST endpoints and our backend automatically handles compilation, execution, and processing of quantum circuits. 3 lines of code."},q5:{question:"What happens if the quantum processor goes down?",answer:"Automatic classical fallback. Every quantum function has a high-quality classical equivalent. Your applications never stop â zero interruption guaranteed."},q6:{question:"How much does Infernex Quantum cost?",answer:"Quantum Pro starts at $1,999/month (shared QPU, unlimited QRNG, 10M tokens). QPU Enterprise at $27,440/month (dedicated 72-qubit processor, unlimited tokens, 99.99% SLA). Both include a 14-day money-back guarantee."},q7:{question:"Which industries benefit the most?",answer:"Finance & Banking (portfolio optimization), Pharma & Biotech (drug discovery), Defense & Security (post-quantum cryptography), Crypto & HFT (quantum algorithmic trading), and Insurance (risk modeling)."},q8:{question:"Is this real quantum hardware?",answer:"Yes. Infernex Quantum uses a physical 72-qubit superconducting quantum processor. Results come from real quantum gate operations, not classical simulation."}}},cta:{title:"Ready to go",titleHighlight:"quantum",titleSuffix:"?",description:"Join the companies that have already made the quantum leap with Infernex.",subscribe:"Subscribe now",compare:"Compare with a Quant"},founder:{quote:'"God created man⦠we improve him quantumly."',name:"Ricardo Cadet-Marthe",role:"Founder & CEO, Infernex"},testimonials:{title:"What our",titleHighlight:"Quantum clients say",subtitle:"Industry leaders trust quantum computing for their strategic operations.",items:[{sector:"Finance",quote:"QAOA optimization replaced 45 minutes of computation with 12 seconds. Our portfolio allocation has never been more precise.",name:"Head of Quantitative Research",role:"Nordic Investment Fund",savings:"$2.4M/year saved"},{sector:"Pharma",quote:"Quantum Monte Carlo simulations accelerated our molecular screening by 100x. 3 candidates identified in weeks instead of months.",name:"VP Drug Discovery",role:"European Biotech Lab",savings:"100x faster"},{sector:"High-Frequency Trading",quote:"QAOA decisions in under 50ms on 72 qubits. Our signal generation is now unpredictable thanks to QRNG.",name:"CTO Trading Systems",role:"HFT Firm",savings:"+15% alpha"}]}},Cd={seo:{title:"Quant Analyst vs Infernex Quantum QPU â Cost Comparison 2026",description:"Compare hiring a quant analyst ($200k/year) vs Infernex Quantum QPU ($1,999/month). 92% savings, 24/7 availability, zero human error."},hero:{badge:"⦠COMPARATIVE ANALYSIS",titleQuant:"Quant Associate",titlePrice1:"$200k/year",titleOr:"or Infernex QPU",titlePrice2:"$27,440/month",titleEnd:"?",stats:{jr:{label:"Avg Quant Jr Salary",value:"$125k",source:"Glassdoor 2026"},sr:{label:"Avg Quant Sr Salary",value:"$250k",source:"JPMorgan data"},infernex:{label:"Infernex QPU",value:"$27,440/mo",source:"Public pricing"}}},comparison:{title:"Detailed",titleHighlight:"comparison",criteria:"Criteria",jr:"Associate Jr",sr:"Associate Sr",infernex:"Infernex QPU",rows:{annualCost:{feature:"Annual cost",jr:"$125,000",sr:"$250,000",infernex:"$329,280"},cost3y:{feature:"3-year cost",jr:"$375,000",sr:"$750,000",infernex:"$987,840"},availability:{feature:"Availability",jr:"8h/day",sr:"10h/day",infernex:"24/7/365"},recruitTime:{feature:"Recruitment time",jr:"3-6 months",sr:"6-12 months",infernex:"Instant"},scalability:{feature:"Scalability",jr:"1x",sr:"1x",infernex:"â"},humanErrors:{feature:"Human errors",jr:"Frequent",sr:"Occasional",infernex:"0"},vacation:{feature:"Vacation/Sick leave",jr:"30d/year",sr:"30d/year",infernex:"0"},training:{feature:"Ongoing training",jr:"$15k/year",sr:"$25k/year",infernex:"Auto-update"},models:{feature:"Simultaneous models",jr:"2-3",sr:"5-10",infernex:"Unlimited"},latency:{feature:"Decision latency",jr:"Hours",sr:"Minutes",infernex:"<50ms"},quantum:{feature:"Quantum computing",jr:"No",sr:"Theoretical",infernex:"72 qubits"},bias:{feature:"Cognitive bias",jr:"High",sr:"Moderate",infernex:"0"},coverage:{feature:"Market coverage",jr:"1-2 classes",sr:"3-5 classes",infernex:"All"},roi:{feature:"Average ROI",jr:"2-5x",sr:"5-10x",infernex:"100x+"}}},math:{title:"The",titleHighlight:"math",titleSuffix:"speaks",optionA:{badge:"Option A â Quant Team",price:"$330k/year",desc:"1 Sr Quant ($250k) + payroll taxes (35%) + training ($15k) + tools ($10k)",items:["Recruitment time: 6 months","Attrition risk: high","Scalability: linear (1 hire = 1x)"]},optionB:{badge:"Option B â Infernex QPU",price:"$27,440/mo",desc:"All-inclusive. Dedicated 72-qubit QPU + classical AI + Slack support",items:["Available immediately","Attrition risk: 0","Scalability: infinite"]}},notReplacement:{title:"Not a replacement.",titleHighlight:"A multiplier.",subtitle:"1 quant + Infernex =",subtitleBold:"10 classical quants",yourQuant:{title:"Your quant",desc:"Strategy, intuition, client relationships"},infernex:{title:"Infernex QPU",desc:"Execution, optimization, scalability"}},roiCalc:{title:"Calculate your",titleHighlight:"savings",quantsLabel:"Number of quants on team",salaryLabel:"Average salary ($)",aumLabel:"AUM / Budget ($)",teamCost:"Current team cost",savings:"Savings with Infernex"},faq:{title:"Frequently Asked",titleHighlight:"Questions",items:{q1:{question:"Can Infernex Quantum really replace a quant analyst?",answer:"Not entirely â it augments them. Infernex handles computation, backtesting, and optimization at quantum speed. Your team focuses on strategy, market intuition, and client relationships. Result: better decisions, faster, at 90% lower cost."},q2:{question:"How does the cost compare to hiring a quant?",answer:"A junior quant costs $125kâ$150k/year. A senior quant costs $200kâ$300k/year. Infernex Quantum Pro starts at $1,999/month ($24k/year) â up to 92% savings with 24/7 availability and zero human error."},q3:{question:"What tasks can Infernex Quantum perform vs a human quant?",answer:"Portfolio optimization (QAOA), risk modeling, Monte Carlo simulations (QRNG), algorithmic trading signals, options pricing, and stress testing â all via API, in milliseconds instead of hours."},q4:{question:"Is quantum computing really faster for finance?",answer:"For certain optimization problems (combinatorial, NP-hard), yes. QAOA on 72 qubits can find optimal portfolio allocations that classical methods miss. Our tests show ~2% additional risk-adjusted return on $500M+ portfolios."},q5:{question:"What about regulatory compliance?",answer:"Infernex is SOC 2 Type II and GDPR compliant. Data hosted in Switzerland. Full audit trail. Zero data retention on queries. We meet the same compliance standards as Bloomberg and Refinitiv."},q6:{question:"Is there a trial period?",answer:"Yes. 14-day money-back guarantee on all Quantum plans. If you don't see measurable improvement in your workflows, we refund you 100%."}}},closing:{title:"The choice is",titleHighlight:"obvious",titleEnd:".",description:"Augment your quant team, don't replace it. Infernex QPU is the quantum co-pilot your desk needs.",cta:"View Quantum Plans"},founder:{quote:'"God created man⦠we improve him quantumly."',name:"Ricardo Cadet-Marthe",role:"Founder & CEO, Infernex"}},Pd={navLabel:"⦠Connect",dashboard:{active:"Actif",addSubAccount:"Add Sub-Account",apiKeysDesc:"Manage your White Label API keys. Use them to authenticate your subaccounts.",apiKeysTitle:"API keys",billingTitle:"Facturation & Utilisation",brandDesc:"Customize the look of your White Label integration.",brandName:"Nom de marque",brandTitle:"Personnalisation de marque",currentPlan:"Plan actuel",inactive:"Inactif",month:"/mois",monthlyTokens:"Tokens mensuels",noApiKeys:"No API keys yet. Keys will be generated once your request is approved.",noSubAccounts:"No subaccounts yet. Create your first customer account.",pendingDesc:"We are reviewing your request. You will receive an email once approved.",pendingTitle:"Demande en cours d'examen",primaryColor:"Couleur principale",revenueEarned:"Shared revenue earned",revenueShare:"Revenue Share",saveChanges:"Enregistrer",secondaryColor:"Couleur secondaire",subAccounts:"Sous-comptes",subAccountsDesc:"Manage your customers' accounts under your brand.",subAccountsTitle:"Sous-comptes",tabs:{apiKeys:"API keys",billing:"Facturation",branding:"Personnalisation",overview:"Vue d'ensemble",subAccounts:"Sous-comptes"},thisMonth:"Ce mois-ci",title:"Tableau de bord {{brand}}",tokens:"tokens",website:"Site web"},features:{fullWhiteLabel:"White Label complet",fullWhiteLabelDesc:"Your logo, your colors, your domain. Customers never see Infernex.",global:"Infrastru
132cture mondiale",globalDesc:"Low-latency inference from data centers around the world.",multiTenant:"API multi-tenant",multiTenantDesc:"Create unlimited sub-accounts for your customers with isolated usage tracking.",revenue:"Partage de revenus",revenueDesc:"Earn up to 30% of your customers' usage revenue.",security:"Business security",securityDesc:"SOC 2 compliant infrastructure with dedicated API keys per partner.",setup:"Instant setup",setupDesc:"Get started in minutes with our simple API integration."},form:{badge:"Enterprise uniquement",brandName:"Nom de marque *",companyName:"Nom de l'entreprise *",contactEmail:"Email de contact *",contactName:"Nom du contact *",submit:"Request an Enterprise quote",submitting:"Envoi en cours...",subtitle:"For custom needs, dedicated infrastructure or unlimited usage - contact our team.",title:"Partenariat Enterprise",useCase:"Describe your business needs",useCasePlaceholder:"Describe your expected volume, custom requirements, SLA needs...",websiteUrl:"URL du site web"},hero:{badge:"Infernex Connect",cta:"Commencer",subtitle:"Like Stripe Connect, but for AI. Offer powerful LLM inference to your clients with your brand, pricing, and keep up to 85% of revenue.",title:"Resell AI inference",titleHighlight:"under your brand"},pricing:{badge:"Tarifs Infernex Connect",contactSales:"Contact Sales",mostPopular:"Most Popular",perMonth:"/mois",startNow:"Commencer â",subtitle:"Choose the plan that suits your business. Evolve as you grow.",title:"Niveaux de partenariat",tokensMonth:"tokens/mois"},seo:{description:"Resell Infernex AI inference under your own brand with Infernex Connect. White label partnership program starting at $299/month with up to 85% revenue.",title:"Infernex Connect - Partenariat White Label IA"},toasts:{checkoutError:"Failed to create payment session",copied:"Copied to clipboard!",loginRequired:"Please log in to apply",loginToSubscribe:"Please log in to subscribe",submitError:"Failed to send request",submitted:"Request sent! We will review it within 24-48 hours."},tiers:{starter:{name:"Starter"},growth:{name:"Growth"},pro:{name:"Pro"},enterprise:{name:"Enterprise"}}},Td={badge:"Proof of Concept",title:"Start Your",titleHighlight:"Free POC Pilot",subtitle:"Test Infernex on your own workloads â zero risk, zero commitment. See real savings in 48 hours.",includedTitle:"What's Included",benefits:["1M free tokens to test any model","Dedicated engineering support","Custom benchmarks on your data","Full performance report with ROI analysis","Seamless migration assistance"],timeline:"Timeline",steps:["Day 1 â Kick-off call & API setup","Day 2 â Run your workloads on Infernex","Day 3 â Receive performance & savings report"],formTitle:"Request Your Free POC",formDesc:"Fill in the form and our team will reach out within 24 hours.",namePlaceholder:"Your name",emailPlaceholder:"Work email",companyPlaceholder:"Company name",useCasePlaceholder:"Describe your use case (optional)",submitBtn:"Request POC",noCommitment:"No credit card required · No commitment",successTitle:"POC Request Received!",successDesc:"Our team will c
132ontact you within 24 hours to schedule your kick-off call.",successToast:"POC request submitted successfully!",errorToast:"Something went wrong. Please try again."},Ed={steps:{useCase:{title:"Use Case",description:"Tell us what you'll build"},volume:{title:"Volume",description:"Estimate your usage"},signup:{title:"Create Account",description:"Get started in minutes"}},step1:{title:"Download",description:"Select your primary use case so we can tailor your experience.",download_mac_arm:"Download for Mac (Apple Silicon)",download_mac_intel:"Download for Mac (Intel)",download_win:"Download for Windows",download_linux:"Download for Linux",advanced_link:"Advanced installation (developers)"},step2:{title:"Install & Launch",description:"Pick the volume that best matches your needs. You can upgrade anytime.",desc:"Follow the visual instructions to install. No terminal required.",detecting:"Detecting agent...",agent_detected:"Agent detected!",waiting:"Waiting for the agent to start..."},step3:{description:"Your recommended plan is {{plan}}. Create your account to get started.",title:"Create Your Token",desc:"Generate a secure token to protect your local agent.",generate:"Generate Secure Token",token_copied:"Token copied to clipboard",token_weak:"Weak token",token_strong:"Strong token",token_excellent:"Excellent token"},useCases:{chatbot:{label:"Chatbot / Assistant",tokens:"~2M tokens/mo"},translation:{label:"Translation",tokens:"~5M tokens/mo"},analysis:{label:"Data Analysis",tokens:"~10M tokens/mo"},coding:{label:"Code Generation",tokens:"~8M tokens/mo"},gaming:{label:"Gaming AI",tokens:"~15M tokens/mo"},enterprise:{label:"Enterprise Platform",tokens:"~50M+ tokens/mo"}},volumes:{starter:{label:"Up to 500K tokens/mo",plan:"Starter",price:"Free"},developer:{label:"Up to 5M tokens/mo",plan:"Developer",price:"$29/mo"},pro:{label:"Up to 50M tokens/mo",plan:"Pro",price:"$149/mo"},enterprise:{label:"Unlimited tokens",plan:"Enterprise",price:"Custom"}},continue:"Continue",back:"Back",plan:"Recommended:",savingsEstimate:"You could save ~${{amount}}/mo vs OpenAI",savingsDescription:"Based on your estimated volume, switching to Infernex could significantly reduce your AI inference costs.",continueWithGoogle:"Continue with Google",connecting:"Connectingâ¦",orByEmail:"or sign up with email",email:"Email",emailPlaceholder:"[email protected]",password:"Password",passwordPlaceholder:"Min. 8 characters",secureData:"Data encrypted",apiIn2Min:"API key in 2 min",noCommitment:"No commitment",createAccount:"Create Account",creating:"Creatingâ¦",alreadyHaveAccount:"Already have an account?",signIn:"Sign in",error:"Error",fillAllFields:"Please fill in all fields.",accountCreated:"Account created!",checkEmail:"Check your email to verify your account.",subtitle:"Your private AI agent, ready in 5 minutes",step4:{title:"Test Connection",test_btn:"Test Connection",success:"Connection successful!",fail_not_running:"Agent is not running. Please start it first.",fail_wrong_token:"Invalid token. Please regenerate.",fail_port_blocked:"Port blocked. Check your firewall settings.",enter_dashboard:"Enter Dashboard"},step5:{title:"Setup Complete!",skip:"Skip",configure_later:"Configure channels later"}},Rd={seo:{title:"Infernex Photonique â Green Photonic AI Inference",description:"Photonic AI inference with zero energy, <10µs latency. All-in-One bundle: AI + Photonique + Quantum 72-qubit. 95% cheaper."},hero:{badge:"Infernex Quantum â 72 qBit Quantum Brain",title:"Choose Infernex Photonique,",titleHighlight:"Infernex Package All-in-One",subtitle:"The only SaaS in the world with AI inference + green photonics + 72 qBit quantum brain. 95% cheaper than everywhere else. Fully automatic.",quantumBrain:"Infernex Quantum is the 72 qBit quantum brain powering this entire page"},loginRequired:"Please log in to subscribe",plans:{title:"Choose Your Photonic Plan",subtitle:"Powered by the Infernex Quantum 72 qBit quantum brain. All prices dynamically converted to your local currency.",private:"Private",enterprise:"Enterprise",month:"mo",year:"yr",or:"or",cta:"Choose This Plan",ctaYearly:"Save 17% â Annual Billing",greenStar:{name:"Infernex Photonique (Green Star)",desc:"Pure photonic inference (ONN), zero energy, latency < 10 µs.",features:["Photonic neural network (ONN)","Near-zero energy consumption","Latency < 10 µs","OpenAI-compatible API","Green carbon footprint","Auto-routing by quantum brain"]},allInOne:{name:"Infernex All-in-One Bundle",desc:"AI Inference + Photonique + Quantum in a single subscription. Shared quantum brain (based on Quantum Pro at $1,990).",badge:"Ultimate: AI + Photonique + Quantum (shared mode)",features:["Everything in Photonique Green Star","Full AI Inference (GPT-4o, Claude, Llamaâ¦)","72-qubit Quantum brain (shared mode)","QAOA portfolio optimization","QRNG true random generation","Priority support & SLA 99.9%"]},monthly:"Monthly",annual:"Annual",ctaEnterprise:"Choose Enterprise Plan"},comparison:{title:"Detailed Comparison",feature:"Feature",rows:[{label:"Energy",photonique:"Near-zero (photonic)",quantum:"Low (cryogenic)",allInOne:"â
Near-zero + Low"},{label:"Latency",photonique:"< 10 µs",quantum:"< 50 ms",allInOne:"â
< 10 µs (photonic route)"},{label:"Token Price",photonique:"$0.03/M",quantum:"$0.10/M",allInOne:"â
From $0.03/M"},{label:"Precision",photonique:"99.5%",quantum:"99.2%",allInOne:"â
99.5%"}
132,{label:"Quantum Brain Level",photonique:"N/A",quantum:"Dedicated 72 qBit",allInOne:"â
Shared 72 qBit ($1,990 plan)"},{label:"Green Impact",photonique:"â
Zero carbon",quantum:"â¡ Low",allInOne:"â
Zero carbon (photonic)"},{label:"QAOA Optimization",photonique:"â",quantum:"â
Full",allInOne:"â
Shared mode"},{label:"QRNG Random",photonique:"â",quantum:"â
Full",allInOne:"â
Shared mode"}],allInOne:"All-in-One"},whyUnbeatable:{title:"Why These Prices Are Unbeatable",text:"Photonic technology is mature and scalable: Optical Neural Networks (ONN) process data at the speed of light with near-zero electricity. Combined with Infernex Quantum in shared mode (based on the $1,990 plan), you get quantum power at a fraction of the dedicated cost.",point1:"Photonics: speed of light, zero heat",point2:"Quantum shared: 72 qBit at reduced cost",point3:"Quantum brain auto-routes your models"},api:{title:"Photonic API Infernex â The Most Powerful on the Market",subtitle:"With Infernex Photonique, you get green, ultra-fast AI inference (latency < 10 µs) and up to 95% less energy consumption. The Infernex quantum brain automatically routes your models to photonics once your plan is active.",keyTitle:"Get Your API Key in 1 Click",keyText:"Access to the API (OpenAI-compatible) and personal Dashboard is granted automatically after subscription only. In your Dashboard you visualize in real-time all your inference data, quantum or photonic.",getCta:"Get My Photonic API Key"},payment:{title:"Secure & Automatic Payment",text:"The Infernex Quantum brain routes your model automatically and bills to the cent. Stripe-secured payment."},cta:{title:"Join the Photonic & Quantum Evolution",subtitle:"1 unique SaaS
132in the world: Infernex AI. Inference + Photonics + Quantum. All powered by the 72 qBit quantum brain.",button:"Start Now"},explainer:{title:"What is photonic AI inference?",intro:"Photonic inference replaces electrons with photons: matrix multiplications â the core of every LLM â are executed at the speed of light directly inside optical chips (ONN, Optical Neural Networks). No transistor switching, no thermal loss, no memory back-and-forth. A revolution comparable to the shift from vacuum tubes to silicon.",usageTitle:"Concrete use cases",usages:["Massive LLM inference (chatbots, RAG, agents) at 1/10th the cost per million tokens","Real-time computer vision (autonomous vehicles, industrial quality control, medical imaging)","High-frequency trading & risk scoring (< 1 ms photonic latency)","Voice AI & real-time translation at scale for call centers","Genomics, drug discovery and climate simulation (heavy tensor workloads)"],gainsTitle:"Proven gains for enterprises",gains:["â 90% less energy consumed per inference vs. an NVIDIA H100 GPU (0.3 W vs. 3â4 W per equivalent op)","Up to 10Ã cheaper per million tokens: $0.20 vs. $2.00 on a classic GPU stack","100Ã lower latency on matrix ops â photons travel at c, no clock cycles wasted","â 95% COâ reduction: a datacenter of 10,000 photonic chips = 300 MWh/year vs. 6,000 MWh/year for GPUs","Real case: a Fortune 500 customer saves $1.8M/year moving 50M daily inferences from GPU to photonic"],revolution:"This is not an incremental upgrade â it is a physical rupture. Photonic inference is the only technology capable of scaling AI to 10 billion daily users without collapsing the planet's energy grid."}},Od={posts:{hardwareVsApi:{seoTitle:"Hardware vs API Keys: Why Infernex Wins Without GPU",seoDescription:"TensorRT, OpenVINO, ONNX â impressive tech, but not for everyone. Discover how Infernex delivers equivalent performance without any GPU investment.",title:"Hardware vs API Keys: Why Infernex Wins Without GPU",subtitle:"TensorRT, OpenVINO, ONNX â impressive technologies... but not for everyone.",intro:{title:"Introduction: The Race for Fast Inference",p1:"In recent years, AI inference has become a technological battlefield. On one side, hardware giants like NVIDIA and Intel develop ultra-optimized engines to run models at incredible speeds. On the other, a new generation of platforms proves it's possible to achieve the same performance â or even surpass it â without investing a cent in physical infrastructure.",p2:"Infernex belongs to this second category. No GPU. No datacenter. No DevOps team. Just an intelligent approach to inference via optimized API keys."},section1:{title:"1. Hardware Inference Engines: Powerful but Costly",tensorrt:{desc:"TensorRT is NVIDIA's flagship inference engine. It optimizes neural networks for NVIDIA GPUs using advanced techniques: quantization (FP32 â INT8), layer fusion, and hardware-specific kernels.",points:["Extremely low on-site latency","Requires dedicated NVIDIA GPUs","DevOps team required for maintenance","Initial cost often exceeds $100,000"]},openvino:{desc:"OpenVINO is Intel's equivalent for its own architectures (CPU, integrated GPU, VPU). It's open source and particularly suited for edge deployments.",points:["Excellent for edge computing on Intel hardware","Open source (Apache 2.0 license)","Less relevant for large-scale cloud APIs"]},onnx:{desc:"ONNX (Open Neural Network Exchange) is the most universal of the three. It acts as a bridge between different frameworks (PyTorch, TensorFlow) and different hardware backends.",points:["Multi-platform by design","Open source (MIT license)","Used in production by Microsoft, Meta, and others"]},summary:"These three technologies are powerful, but they require you to manage your own hardware infrastructure. It's a massive investment in time, money, and skills."},section2:{title:"2. The Infernex Approach: Performance Without Hardware",desc:"Infernex takes a radically different philosophy. Instead of managing expensive GPUs, we intelligently aggregate the best AI providers â OpenAI, Anthropic, Mistral, Groq â and expose them via a unified, OpenAI-compatible API.",points:["ð Intelligent routing to the fastest provider per request","ð Multi-region infrastru
132cture â latency < 5ms","â¡ Access to the fastest providers (Groq LPU, etc.)","ð° Up to 95% cheaper through volume aggregation"]},section3:{title:"3. Direct Comparison: Hardware vs Infernex"},table:{criteria:"Criteria",cost:"ð° Initial Cost",latency:"â¡ Latency",scalability:"ð Scalability",maintenance:"ð§ Maintenance",multiModel:"ð Multi-Models",updates:"ð Updates",costPerToken:"ðµ Cost/M Tokens",limited:"Limited to hardware",unlimited:"Unlimited, instant",devopsRequired:"DevOps team required",none:"None",difficult:"Difficult to change",manual:"Manual, costly",automatic:"Automatic",from005:"From $0.05"},section4:{title:"4. When Hardware Remains Relevant",desc:"Let's be honest: there are use cases where TensorRT or OpenVINO remain the best option:",points:["Fully offline deployments (no internet access)","Regulatory constraints requiring data to never leave the site","Astronomical volumes (billions of requests/day) justifying the investment","Real-time computer vision on industrial cameras (Ultralytics/YOLO)"],conclusion:"For these very specific cases, dedicated hardware can be justified. But for the vast majority of startups, SMBs, and developers, it's a disproportionate investment."},section5:{title:"5. Conclusion: Intelligence Over Brute Force",p1:"TensorRT, OpenVINO, and ONNX are remarkable technologies that have revolutionized AI inference for those who can afford to deploy them. But they represent a past approach: investing massively in infrastructure to control every layer of the stack.",p2:"Infernex represents the modern approach: leveraging the best providers on the market, orchestrating them intelligently, to deliver equivalent performance at a fraction of the cost. No GPU. No maintenance. No surprises."},cta:{title:"Try Infernex for Free",desc:"Deploy your first model in under 5 minutes on infernex.ai â from $0.05/M tokens.",button:"Start Free Now"}},cheapInference:{altHardware:"Hardware alternatif",altHardwareText:"The AMD MI300X, Intel Gaudi 2, and AWS Inferentia offer performance comparable to NVIDIA at 40-60% of the cost. Infernex uses these alternatives to maximize savings.",avgReduction:"average reduction",batching:"Batching intelligent",batchingText:"Consolidating multiple queries into a single GPU execution can reduce costs by 50-70% for non-real-time use cases.",caching:"Caching and Cache Management",cachingText:"Storing frequent responses can reduce API calls by 30-50% depending on use cases.",conclusion:"Conclusion",conclusionText:"Affordable AI inference is no longer a dream. With the right techniques and platforms, you can reduce your costs by 70-95% while maintaining quality comparable to market leaders.",ctaButton:"Free Trial",ctaText:"Try Infernex for free â 50K tokens/month included, no card required.",ctaTitle:"Ready to reduce your AI costs?",description:"Reduce your AI inference costs by 70-95% without compromising performance.",intro:"Reduce your inference costs by 70-95% with modern optimization techniques, alternative hardware, and specialized platforms like Infernex.",openSource:"Optimized open-source models",openSourceText:"Llama 3, Mistral, and Qwen rival GPT-4 for most tasks at a fraction of the cost.",quantization:"Quantization (4-8 bit)",quantizationText:"Quantization reduces the precision of model weights from 32 bits to 4-8 bits, decreasing memory requirements and speeding up inference by 2-4x with minimal quality loss (<1%).",savedFor1B:"saved for 1B tokens/year",summary:"Summary: Potential savings",techniques:"The 5 techniques to reduce costs",title:"Cheap AI Inference: Complete Guide 2026",titleHighlight:"Cheap",vsMarket:"vs $2-8 market",whyExpensive:"Why is AI inference so expensive?",whyExpensiveText:"Large providers like OpenAI, Anthropic and Google charge between $2 and $8 per million tokens. These high prices can be explained by several factors: expensive GPU infrastructure (NVIDIA H100 at $30,000 per unit), significant commercial margins, and lack of competition until recently."},economicLLM:{after:"After Infernex migration",afterText:"$180k/an",before:"Avant",beforeText:"$2.4M/year with OpenAI",caseStudy:"Case study: FinTech 500 employees",check1:"Audit your current use cases and token volumes",check2:"Identify the appropriate models for each task",check3:"Calculate the ROI
132with our savings simulator",check4:"Deploy a POC on a non-critical use case",check5:"Gradually migrate other applications",checklist:"Enterprise deployment checklist",contactEnterprise:"Contact the Enterprise team",costProblem:"The problem of the cost of LLMs in business",costProblemText:"A company processing 100 million tokens per day with GPT-4 spends around $3M per year. This prohibitive cost prevents many organizations from benefiting from generative AI.",ctaText:"Our Enterprise team supports you in analyzing your needs and calculating your ROI.",ctaTitle:"Calculate your potential savings",description:"Deploy LLMs in companies with a positive ROI. Case studies and real savings.",intro:"LLMs are no longer reserved for tech giants. Learn how businesses of all sizes are deploying AI with positive ROI.",migrationTime:"Temps de migration",migrationTimeText:"2 heures (changement d'URL API)",modelSelection:"Choosing the appropriate model",modelSelectionText:"GPT-4 is not needed for 80% of tasks. Optimized models like Llama 3 70B or Mistral Large offer comparable performance.",optimizedInfra:"Optimized infrastructure",optimizedInfraText:"Using alternative hardware (AMD, Intel) and techniques like batching and caching can reduce costs by 60-80%.",savingsResult:"Savings",seePricing:"See pricing",solution:"The solution: Multi-level optimization",specializedPlatforms:"Specialized platforms",specializedPlatformsText:"Infernex combines all these optimizations to offer tokens starting at $0.05/M with a latency <10ms and a 100% OpenAI compatible API.",title:"Economic LLM for Businesses: ROI Guide",titleHighlight:"Economical",useCasesTitle:"Use cases and real savings",customerSupport:"Automated Customer Support",customerSupportDesc:"80% reduction in support costs with optimized LLM chatbots",docAnalysis:"Document Analysis",docAnalysisDesc:"Processing 10M documents/month at $0.05/M tokens vs. $2/M",moderation:"Content Moderation",moderationDesc:"Real-time moderation with latency <10ms guaranteed",codeGen:"Code Generation",codeGenDesc:"Large-scale development wizards for teams"},openaiAlternative:{claudeText:"Great for complex reasoning tasks, but different API from OpenAI (migration required) and still high prices.",ctaButton:"Commencer gratuitement â 50K Tokens/mois",ctaText:"API 100% compatible OpenAI. Aucune modification de code requise.",ctaTitle:"Migrate in 2 minutes, save 95%",description:"Complete comparison of alternatives to OpenAI: price, latency, API compatibility.",geminiText:"Good value for money for multimodal, but Google ecosystem and proprietary API.",highPrices:"High prices",highPricesText:"GPT-4 costs $30/M tokens, inaccessible for scaling",infernexBest:"Infernex : L'alternative #1 en 2026",infernexBestText:"Infernex stands out for its 100% OpenAI compatibility (just change the URL), its ultra-low latency (<10ms), and its prices 95% lower than the market.",intro:"OpenAI is no longer the only option. Discover alternatives that offer equivalent performance at 70-95% cheaper.",latency:"Latence",migration2min:"Migration en 2 minutes",openaiCompatible:"Compatible OpenAI",otherAlternatives:"Autres alternatives notables",pricePerM:"Prix/M tokens",provider:"Fournisseur",rateLimits:"Limites de rate",rateLimitsText:"Restrictions on query volume",savings:"Savings",selfHostedText:"Open-source models perform well, but require expensive GPU infrastructure and DevOps expertise.",title:"Best OpenAI Alternatives in 2026",titleHighlight:"Alternatives to OpenAI",variableLatency:"Latence variable",variableLatencyText:"200-500ms en pic d'utilisation",vendorLockin:"Vendor lock-in",vendorLockinText:"Costly migration to other providers",whyAlternative:"Why look for an alternative to OpenAI?"}},switchOpenai:{seoTitle:"Why We Switched from OpenAI to Infernex and Saved $7,600/month",title:"Why We Switched from OpenAI to Infernex and Saved $7,600/month",meta:"A real case study: how a Series A startup cut their OpenAI bill from $8,000 to $400/month by changing one line of code.",category:"Case Study",readTime:"6 min read",intro:"We were spending $8,000 per month on OpenAI. Our AI features were growing, our costs were exploding, and our investors were asking questions. The
132n we found Infernex.",s1Title:"The Problem â Our OpenAI Bill Was Exploding",s1Body:"As a Series A startup building AI-powered analytics, we relied heavily on GPT-4o for text analysis, summarization, and code generation. At first, the costs were manageable â a few hundred dollars per month. But as we scaled to 50,000+ daily API calls, our monthly OpenAI bill hit $8,000 and was growing 30% month-over-month. Our runway was shrinking faster than our revenue was growing.",s2Title:"The Discovery â Finding Infernex",s2Body:"Our CTO discovered Infernex on Hacker News. The promise seemed too good to be true: the same OpenAI-compatible API, the same models, but 95% cheaper. We were skeptical. But at $8,000/month, we had nothing to lose by testing it.",s3Title:"The A/B Test â 2 Weeks, 50% of Traffic",s3Body:"We ran a rigorous A/B test. 50% of our traffic went through OpenAI, 50% through Infernex. Same prompts, same models, same everything. The only difference was the base_url. Here is the entire migration:",s3Code:`// Before (OpenAI) 133const client = new OpenAI({ apiKey: "sk-..." }); 134 135// After (Infernex) â literally 1 line changed 136const client = new OpenAI({ 137 apiKey: "infx-...", 138 baseURL: "https://api.infernex.ai/v1" 139});`,s4Title:"The Results â Identical Quality, 91% Cheaper",s4Body:"After 2 weeks and 700,000 API calls, the results were clear. Quality: our internal quality score showed 99.3% similarity between OpenAI and Infernex responses. Latency: Infernex averaged 4.2ms vs OpenAI's 380ms â 90x faster. Cost: $368 on Infernex vs $4,100 on OpenAI for the same volume. We switched 100% of our traffic to Infernex the next day.",s5Title:"The Migration â 3 Lines of Code",s5Body:"The entire migration took 12 minutes. We changed the base URL, swapped the API key, and deployed. Zero downtime. Zero broken features. Zero customer complaints.",s5Code:`// pip install openai (same SDK) 140import openai 141 142client = openai.OpenAI( 143 api_key="infx-your-key", 144 base_url="https://api.infernex.ai/v1" 145) 146 147# That's it. Everything else stays the same.`,s6Title:"Six Months Later â The Impact",s6Body:"Six months after the switch, we have saved $45,600 in total API costs. Our latency improved from ~400ms to ~5ms, which directly improved our user experience metrics. Our investors are happy. Our users are happy. And we extended our runway by 4 months â just from this one infrastructure decision.",ctaTitle:"Ready to cut your AI costs by 95%?",ctaBody:"Start free. Same code. Same models. 95% cheaper.",ctaBtn:"Start Free Today"},badge:"Blog & Guides",title:"AI Inference",titleHighlight:"Guides",subtitle:"Practical guides to reduce your AI costs and optimize your infrastructure.",filterByTag:"Filter by tag",allPosts:"All posts",readMore:"Read full article",loading:"Loading articles...",noArticles:"No articles found for this tag.",articleNotFound:"Item not found",backToBlog:"Retour au blog",ctaButton:"Start free",ctaText:"Try Infernex free â 50K tokens/month included, no card.",ctaTitle:"Ready to cut your inference costs by 95%?",linkCopied:"Link copied!",readTime:"read",share:"Partager :",views:"vues",relatedArticles:"Related Articles",articles:{azureVsInfernex:{title:"Azure AI Costs vs Infernex â Save 95%",subtitle:"Microsoft recommends FinOps and provisioned throughput to manage Azure AI costs. But what if you could skip the complexity and save 95% by switching your base URL?",costProblem:"The Azure AI Cost Problem",costProblemP1:`According to Microsoft's own blog, AI cloud costs "can scale as quickly as workloads." Their recommended solution? Provisioned throughput reservations, FinOps strategies, and careful budget management.`,costProblemP2:"But there's a simpler approach: route your inference through Infernex and pay up to 95% less â with zero infrastructure changes.",costComparison:"Cost Comparison: Azure AI vs Infernex",model:"Model",azurePrice:"Azure OpenAI (per 1M tokens)",infernexPrice:"Infernex (per 1M tokens)",savings:"Savings",migration:"Migration: 2 Lines of Code",whyBetter:"Why Infernex Beats Azure FinOps",noProvisioned:"No provisioned throughput â pay only for what you use",noLockin:"No vend
147or lock-in â standard OpenAI SDK compatibility",autoRouting:"Automatic routing â best model selected per request",noFinops:"No FinOps team needed â costs are simply 95% lower",keyTakeaways:"Key Takeaways",avgReduction:"average cost reduction vs Azure AI",linesToMigrate:"of code to migrate",avgLatency:"average latency",ctaTitle:"Stop Overpaying for Azure AI",ctaText:"Get 50K free tokens/month and see the savings for yourself.",ctaButton:"Start Free â 50K Tokens/month"},azure70:{title:"Cut Azure AI Costs by 70%? Try 95% with Infernex",subtitle:"A viral Medium article explains how to save 70% on Azure AI while scaling GPT-5. But what if 70% isn't enough? Here's how to hit 95%.",mediumApproach:"The Medium Article's Approach",mediumP1:'In "How to Cut Azure AI Costs by 70%", the author recommends techniques like:',ptu:"Provisioned throughput units (PTUs) for predictable workloads",promptOpt:"Prompt optimization and token reduction",modelSelection:"Model selection (choosing smaller models when possible)",batchProcessing:"Batch processing for non-real-time tasks",mediumP2:"These are sound practices. But they still keep you on Azure's pricing tier and require significant engineering effort.",numbersTitle:"70% vs 95%: The Numbers",monthlySpend:"Monthly spend",noOpt:"Azure (no optimization)",azure70col:"Azure (-70%)",infernex95col:"Infernex (-95%)",scaleP:"At enterprise scale, the difference between 70% and 95% savings is $50,000/month.",whyFurther:"Why Infernex Goes Further",diffArch:"Different architecture â routing to open models, not just optimizing Azure consumption",noPtu:"No PTU commitments â pure pay-per-use pricing",autoModel:"Automatic model selection â no manual prompt engineering needed",sdkCompat:"OpenAI SDK compatible â migrate in minutes, not weeks",extra25:"The Extra 25%",extra25Monthly:"additional savings at enterprise scale",extra25Annual:"annual difference (70% vs 95%)",extra25Ptu:"commitments required",ctaTitle:"Don't Settle for 70%",ctaText:"See the 95% difference with 50K free tokens/month.",ctaButton:"Start Free â 50K Tokens/month"},googleCloud:{title:"Google Cloud AI Costs vs Infernex â Why Routing Beats FinOps",subtitle:"Google Cloud Consulting published 3 strategies for optimizing AI costs. We break them down and show why intelligent inference routing is the 4th â and simplest â strategy.",strategies:"Google's 3 AI Cost Strategies (And Their Limits)",strategiesP1:'In their article "Three Proven Strategies for Optimizing AI Costs", Google Cloud recommends:',rightSizing:"Right-sizing infrastructure â choosing the correct GPU/TPU tier",finops:"FinOps practices â monitoring, budgets, cost allocation",committed:"Committed use discounts â 1-3 year reservations for 40-60% savings",strategiesP2:"These are valid approaches â but they require dedicated DevOps teams, infrastructure expertise, and long-term commitments. There's a simpler way.",strategy4:"Strategy #4: Intelligent Inference Routing",strategy4P1:"Infernex analyzes each prompt in real-time and routes it to the optimal open-source model (Llama, Mistral, Mixtral). The result? 95% lower costs with zero infrastructure management.",comparison:"Cost Comparison: Vertex AI vs Infernex",metric:"Metric",vertexCol:"Google Vertex AI",infernexCol:"Infernex",costRow:"Cost per 1M tokens (GPT-4 equiv.)",setupRow:"Setup complexity",setupVertex:"High (GCP project, IAM, VPC)",setupInfernex:"2 lines of code",lockinRow:"Vendor lock-in",lockinVertex:"High",lockinInfernex:"None (OpenAI SDK)",finopsRow:"FinOps team required",yes:"Yes",no:"No",migrationTitle:"Migration from Vertex AI",bottomLine:"Bottom Line",bl1:"savings vs Vertex AI pricing",bl2:"infrastructure to manage",bl3:"migration time",ctaTitle:"Ditch the FinOps Overhead",ctaText:"Try Infernex free with 50K tokens/month â no GCP project needed.",ctaButton:"Start Free â 50K Tokens/month"},benchmark:{title:"AI Inference Cost-Performance Benchmark 2026",subtitle:"Google Cloud's Baseten case study shows 225% better cost-performance with GPU optimization. We compare this approach with Infernex's zero-infrastru
147cture routing.",baseten:"The Baseten Case Study",basetenP1:"In their blog post, Google Cloud showcases how Baseten achieved 225% better cost-performance for AI inference by optimizing GPU allocation on GKE with custom autoscaling.",basetenP2:"This is impressive â but it requires deep Kubernetes expertise, custom HPA configurations, and ongoing infrastructure management. Not every team has those resources.",twoApproaches:"Two Approaches to Cost-Performance",criteria:"Criteria",googleBaseten:"Google Cloud + Baseten",costGain:"Cost-performance gain",setupTime:"Setup time",setupGoogle:"Weeks (K8s, HPA, GPU)",setupInfernex:"5 minutes",teamRequired:"Team required",teamGoogle:"MLOps + DevOps",teamInfernex:"Any developer",maintenance:"Ongoing maintenance",maintenanceHigh:"High",maintenanceNone:"None",lockin:"Vendor lock-in",lockinGke:"GKE + GCP",lockinNone:"None",howTitle:"How Infernex Achieves Superior Cost-Performance",how1:"Intelligent routing â each prompt is analyzed and sent to the most cost-efficient model",how2:"Open-source models â Llama 3, Mistral, Mixtral at fraction of proprietary costs",how3:"Global edge network â <50ms latency without managing any GPUs",how4:"Automatic scaling â no HPA tuning, no node pools, no GPU quotas",summary:"Benchmark Summary",bl1:"lower cost than GKE-managed inference",bl2:"GPUs to manage",bl3:"uptime SLA",ctaTitle:"Skip the GPU Management",ctaText:"Get better cost-performance with zero infrastructure. 50K tokens/month free.",ctaButton:"Start Free â 50K Tokens/month"},semanticCache:{title:"Optimise LLM Usage Costs with Semantic Cache",subtitle:"How caching semantically similar prompts can slash your AI inference bill by up to 90% â and why Infernex includes it natively in Enterprise.",inspiredBy:"Inspired by the original HackerNoon article by Birendra",costProblem:"The Cost Problem with LLM APIs",costP1:"Every API call to GPT-5, Claude, or Gemini costs money. At scale, companies processing millions of requests daily find that 30-60% of prompts are semantically identical or near-identical to previous requests. Customer support bots, FAQ systems, code assistants â they all see massive prompt repetition.",costP2:`Traditional caching (exact string matching) catches only a fraction of these duplicates. A user asking "What's the weather in Paris?" and "Tell me Paris weather" would trigger two separate API calls, despite being semantically identical.`,realNumbers:"The Real Numbers",gpt5Input:"GPT-5 input: $2.50/1M tokens â 1M daily calls = $2,500/day wasted on duplicates",avgHitRate:"Average cache hit rate: 35-55% for production workloads",monthlySavings:"Potential monthly savings: $25,000-$40,000 for a mid-size deployment",howWorks:"How Semantic Caching Works",howP1:"Unlike exact-match caching, semantic caching uses vector embeddings to understand the meaning of prompts. Here's the flow:",step1:"Embed the prompt: Convert the incoming prompt into a vector embedding using a lightweight model (e.g., text-embedding-3-small).",step2:"Search the cache: Query a vector database (Redis, pgvector, Pinecone) for similar embeddings above a similarity threshold (typically 90%).",step3:"Cache HIT: If a match is found, return the cached response instantly â no LLM call needed. Latency drops from ~2s to <10ms.",step4:"Cache MISS: If no match, call the LLM normally, then store both the embedding and the response for future matches.",threshold:"Similarity Threshold: The Critical Parameter",thresholdP1:'The threshold determines how "similar" two prompts must be to trigger a cache hit:',thresh95:"95% â Very strict. Only near-identical prompts. Low false-positive risk but fewer cache hits.",thresh90:"90% â Balanced. Recommended for most production workloads. Good hit rate without sacrificing accuracy.",thresh85:"85% â Aggressive. Higher hit rates but risk serving slightly wrong responses.",implTitle:"Implementation Approaches Compared",approach:"Approach",latency:"Latency",accuracy:"Accuracy",complexity:"Complexity",exactHash:"Exact hash matching",exactHashAcc:"100% (but low hit rate)",low:"Low",redisVector:"Redis + vector search",high:"High (90%+ similarity)",medium:"Medium",pgvector:"pgvector (PostgreSQL)",infernexRow:"Infernex (hash-first, built-in)",infernexAcc:"Exact match + model-aware",zero:"Zero (included)",codeExample:"Code Example: Redis Semantic Cache (DIY)",codeNote:"This requires managing Redis infrastru
147cture, embedding costs, and cache invalidation logic. With Infernex, this is all handled automatically.",infernexTitle:"How Infernex Does It â Zero Configuration",infernexP1:"Infernex's semantic cache is built into the inference proxy. No Redis to manage, no embedding pipeline to build, no similarity thresholds to tune. It works out of the box for Enterprise customers.",hoodTitle:"How it works under the hood:",hood1:"Hash-first strategy: Every request is hashed deterministically based on the model + messages. An exact hash match returns the cached response in <1ms.",hood2:"Model-aware caching: The cache is segmented by model. A GPT-5 response won't be served for a Gemini request, even if the prompt is identical.",hood3:"Automatic TTL: Cache entries expire after 24 hours by default, ensuring freshness without manual invalidation.",hood4:"Transparent headers: Every response includes X-Infernex-Cache: HIT or MISS, so you always know when the cache saved you tokens.",freeEnterprise:"Included Free in Enterprise",freeEnterpriseP:"While other providers charge $99-299/mo for caching add-ons, Infernex includes semantic caching as a core feature of the Enterprise plan. No extra cost. No configuration. Just savings.",migrationTitle:"Migration: 2 Lines of Code",resultsTitle:"Real-World Impact",hitRate:"Avg. Cache Hit Rate",hitRateValue:"of requests served from cache",latencyReduction:"Latency Reduction",latencyReductionValue:"faster on cache hits (<1ms vs 2s)",costSavings:"Cost Savings",costSavingsValue:"on repeated prompt patterns",ctaTitle:"Stop Paying for Repeated Prompts",ctaText:"Infernex's semantic cache is included free in the Enterprise plan. Start saving on your first API call â no Redis, no embeddings pipeline, no configuration.",tryPlayground:"Try the Playground",viewPlan:"View Enterprise Plan"},mirofishInference:{category:"Integrations",readTime:"5 min read",source:"Infernex Quantum Brain",title:"MiroFish + Infernex: Run Up to 500,000 AI Agents for $0.40 per Simulation",intro:"MiroFish is the #1 GitHub trending AI project that builds a digital twin of reality using swarms of AI agents to predict future events. The problem: it is expensive. Each simulation spins up thousands of agents, each making multiple LLM calls. With OpenAI, a single simulation costs $17.50 or more. With Infernex, the same simulation costs $0.40.",whatIs:"What is MiroFish?",whatIsP:"MiroFish is an AI agent swarm engine that reached #1 on GitHub Trending globally. It works by extracting real-world data (news, financial signals, policy documents), generating thousands of AI agents with unique personalities and long-term memory, then simulating their interactions to produce a prediction report. Use cases include: trading signal generation, geopolitical event forecasting, narrative reconstruction, and policy impact analysis.",whyExpensive:"Why MiroFish is expensive by default",whyExpensiveP:"A standard MiroFish simulation involves approximately 2,847 AI agents, scaling up to 500,000 in advanced configurations. Each agent makes 3 to 5 LLM calls during the simulation. A standard run produces approximately 10 million tokens.",howConnect:"How to connect MiroFish to Infernex (3 steps)"},quantum:{backToBlog:"Back to Blog",category:"Quantum",readTime:"12 min",date:"March 8, 2026",title:"Infernex Quantum: Accessible Quantum Computing via Origin Pilot",subtitle:"How we integrated a 72-qubit quantum processor into a REST API accessible in 3 lines of code. And why it changes everything for finance, pharma, and defense.",problem:"The problem: quantum is inaccessible",problemP:"Quantum computing promises exponential gains in optimization, simulation, and cryptography. But today, accessing a quantum processor requires a PhD in physics, a multi-million dollar budget, and months of configuration. Companies that could benefit the most â banks, investment funds, pharmaceutical labs â have neither the time nor the resources to build a dedicated team.",solution:"Our solution: Infernex Quantum",solutionP:"Infernex Quantum integrates a 72-qubit superconducting quantum processor into our existing inference infrastru
147cture. The result: an OpenAI-compatible REST API that automatically routes requests to the QPU when relevant, with an optimized classical fallback when it's not.",qaoa:"QAOA: Quantum Combinatorial Optimization",qaoaP1:"The QAOA algorithm (Quantum Approximate Optimization Algorithm) is the core of our offering. It solves combinatorial optimization problems â portfolio allocation, logistics routing, scheduling â with precision and speed impossible to achieve with classical methods alone.",qaoaP2:"For an investment fund with $500M under management, QAOA can identify optimal allocations that classical Markowitz models don't find. Our internal tests show an average gain of 2% on risk-adjusted return â that's $10M of additional value per year.",qrng:"QRNG: Quantum Cryptographic Security",qrngP1:"Our Quantum Random Number Generator (QRNG) exploits fundamental properties of quantum mechanics to produce truly random entropy. Unlike classical PRNGs which are deterministic, Infernex's QRNG is physically unpredictable.",qrngP2:"Applications: unbreakable encryption keys, unbiased A/B testing, quantum Monte Carlo for options pricing, and seed generation for DeFi smart contracts.",useCases:"Use Cases by Sector",finance:"Finance & Banking",financeUse:"QAOA portfolio optimization, quantum Monte Carlo options pricing, fraud detection",pharma:"Pharma & Biotech",pharmaUse:"Molecular simulation, accelerated drug discovery, protein folding",defense:"Defense & Security",defenseUse:"Post-quantum cryptography, anomaly detection, logistics optimization",crypto:"Crypto & HFT",cryptoUse:"Quantum algorithmic trading, QAOA arbitrage, QRNG for smart contracts",pricing:"Pricing: accessible, not free",pricingP1:"Quantum computing has a real cost. We chose to make it accessible without undervaluing it:",quantumPro:"Quantum Pro â $1,999/month: Shared QAOA, unlimited QRNG, 10M tokens",qpuEnterprise:"QPU Enterprise â $27,440/month: Dedicated 72-qubit QPU, unlimited tokens, 99.99% SLA",pricingP2:"For comparison: Google Willow Quantum is not publicly available. IBM Quantum Platform (Qiskit) starts at $40,000/month. Infernex QPU Enterprise offers real quantum computing for only $27,440/month â the most affordable on the market â with a 14-d
147ay money-back guarantee.",fallback:"Classical fallback: never any downtime",fallbackP:"Quantum is not yet 100% reliable. That's why every Infernex Quantum request has an optimized classical fallback. If the QPU is unavailable or if the problem isn't suited for quantum, we automatically switch to our classical algorithms â with no service interruption, no client-side error.",quote:'"God created man⦠we improve him quantumly."',quoteAuthor:"â Ricardo Cadet-Marthe, Founder & CEO",discoverCta:"Discover Quantum",compareCta:"Compare with a Quant"}}},Md={badge:"Ultimate Supremacy",heroTitle:"Infernex Ultimate Supremacy",heroSubtitle:"TSU PBit â All in One",heroDescription:"Welcome to the world's most advanced compute infrastructure. The Ultimate Supremacy plan fuses four pillars of the digital revolution â AI Inference, Photonics, Quantum, and TSU p-bits â to transform your models into surgical-precision assets achieving 99.7% accuracy. You outperform every competitor. The Holy Grail.",pillarsTitle:"The Alliance of 4 Pillars",pillarsSubtitle:"Why this synergy changes everything",pillars:{photonic:{title:"Speed of Light (Photonics)",description:"The photonic engine handles data transport and processing at sub-nanosecond speeds. By eliminating the electrical resistance of classic chips, photonics enables near-instantaneous information propagation, ideal for massive deep learning layers."},quantum:{title:"Global Optimization (Quantum)",description:"Where standard processors get lost, our quantum processor excels at finding ground states. It scans billions of possibilities to instantly lock the best global structure (the 'Macro')."},tsu:{title:"Precision Sampling (TSU p-bits)",description:"The Thermodynamic Sampling Unit via p-bits is unbeatable for fast stochastic sampling. It doesn't calculate the solution â it lets physics converge towards it via Langevin dynamics. The ultimate tool for uncertainty management (Bayesian AI)."},robustness:{title:"Anti-Overfitting (Robustness)",description:"Controlled thermodynamic noise injection by the TSU acts as a natural regulator. Your AI doesn't just 'repeat' test data; it becomes capable of generalizing on real, unknown data with total reliability."}},perfTitle:"Performance & Precision: Breaking the Last Mile",perfSubtitle:"Reaching 99.2% is a quantum feat. But in AI, each additional 0.1% requires exponential effort. The hybrid architecture breaks this glass ceiling.",table:{metric:"Metric",standard:"Standard (AI + Quantum)",ultimate:"Ultimate (Photonic + Quantum + TSU)",rows:{speed:{label:"Processing Speed",standard:"High",ultimate:"Ultra-Fast (Light Speed)"},precision:{label:"Observed Precision",standard:"99.20%",ultimate:"99.50% to 99.70%"},errorRate:{label:"Residual Error Rate",standard:"0.80%",ultimate:"0.30% to 0.40%"},errorReduction:{label:"Error Reduction",standard:"Baseline",ultimate:"-30% to -50% net reduction"}}},bayesNote:"Why not 100%? There is always a 'Bayes limit' linked to the intrinsic quality of data. However, with the Ultimate Plan, you reach the physical and mathematical frontier of what is possible.",tsuDashTitle:"TSU Precision Dashboard",tsuDashSubtitle:"Watch the TSU clean residual error in real-time using Langevin dynamics and Gibbs sampling.",tsuPrecision:"Precision",tsuResidual:"Residual Error",tsuReduction:"Error Reduction",tsuRunning:"Sampling...",tsuLaunch:"Launch TSU Simulation",techTitle:"Technical Summary",techBitsTitle:"Bit Comparison",techChar:"Characteristic",techClassicBit:"Classic Bit",techQubit:"Qubit (Quantum)",techPbit:"p-bit (Thermo)",techBits:{state:{label:"State",classic:"Fixed (0 or 1)",qubit:"Superposition",pbit:"Probabilistic"},engine:{label:"Engine",classic:"Electrical voltage",qubit:"Quantum coherence",pbit:"Thermal noise / Entropy"},temp:{label:"Temperature",classic:"Ambient",qubit:"Near absolute zero",pbit:"Ambient"}},techGpuTitle:"Why TSU?",techGpu:{method:{label:"Method",gpu:"Matrix multiplication",tsu:"Gibbs Sampling"},energy:{label:"Energy",gpu:"High (Watts)",tsu:"Very low (exploits noise)"},application:{label:"Application",gpu:"Standard Deep Learning",tsu:"Generative AI, Bayesian Inference"}},synergyTitle:"How Synergy Multiplies Your Results",synergy:{cleaning:{title:"Photonic-Quantum Cleaning",description:"Photonics accelerates the flow, quantum defines the trajectory, and TSU (Gibbs sampling) refines local weights with a finesse that deterministic calculation systematically misses."},falsePosElim:{title:"False Positive Elimination",description:"The TSU serves as an intelligent post-processor, cleaning output probabilities to eliminate residual noise that c
147onstitutes the current 0.8% error."},fineTuning:{title:"Thermodynamic Fine-Tuning",description:"Included by default, the TSU intervenes during the final adjustment of your models to capture the last hundredths of precision. Physical sampling surpasses all known software methods."}},closingStatement:"The future is no longer binary. It is Quantum, Photonic, and Thermodynamic. With Infernex.ai, your models are now powered by the fundamental laws of the universe.",planTitle:"Enterprise Plan â Ultimate Supremacy",enterprisePitch:"Why $49,999/mo? Because Ultimate Supremacy bundles OpenAI Enterprise-class inference, Google Quantum AI, and a dedicated photonics lab into one API â at roughly 1/20th of their combined cost, with 99.7% precision and zero vendor lock-in.",form:{name:"Contact Name",company:"Company Name",email:"Corporate Email",corporateOnly:"Infernex Ultimate is reserved for verified Corporate Entities. Please use your professional domain.",requiredFields:"Please fill in all fields.",loginRequired:"Please log in to proceed.",secure:"256-bit encrypted ⢠Enterprise-grade security"},features:{f1:"Multi-routing AI Inference (all models)",f2:"Photonic engine â sub-nanosecond processing",f3:"72-qubit Quantum processor (QAOA)",f4:"TSU p-bit thermodynamic sampling",f5:"Dedicated Enterprise workspace + Supremacy API Key",f6:"Automated onboarding kit â zero human intervention"},cta:"Choose Ultimate Supremacy Plan"}
147,qd={title:"Developers, Startups, Enterprises â choose below the ultimate fusion: your SaaS + Infernex AI Solutions for lifetime savings, lightspeed execution, and 2 unique technologies in 1 API SDK: Quantum & TSU PBit for near 100% accuracy"},Dd={title:"8 AI Tools, One API â 95% Cheaper | Infernex 2026",description:"Cut AI costs by 95%. 8 tools: Inference, Quantum QPU, Photonics, GEO, MiroFish, 3-in-1 Bundle & TSU PBit. From $0.04/M tokens. OpenAI-compatible. Start free."},Ld={lucas:{role:"CTO @ Datalyr (Paris, Series A)",quote:"We were burning $8,000/mo on OpenAI. I changed the base_url on a Friday evening, Monday morning we were at $400. No code touched, no perf degradation. Best infra decision of the year."},rania:{role:"AI Engineer Freelance (Geneva)",quote:"I test LLMs for a dozen clients. Infernex became my default. Latency is indistinguishable from OpenAI, and my margins doubled overnight."},thomas:{role:"Head of Engineering @ Flowsync",quote:"We were afraid of the migration. We ran an A/B test: 50% traffic on Infernex for 2 weeks. Identical quality results, 91% cheaper. We switched 100% the next day."}},Fd={benchmark:{metric:"Metric",priceInput:"Price / 1M tokens input",priceOutput:"Price / 1M tokens output",latency:"Average latency",sdkCompat:"OpenAI SDK compatible",freeTier:"Free tier",sla:"Uptime SLA",note:"Benchmark measured March 2026 â methodology available on request"},proprietarySDK:"Proprietary SDK",switchNow:"Switch to Infernex",anthropic:{meta:{title:"Infernex vs Anthropic Claude â 97% Cheaper AI Inference 2026",description:"Compare Infernex to Anthropic Claude. Same quality, 97% cheaper, 120x faster latency. Drop-in OpenAI-compatible API."},subtitle:"Same Claude-quality outputs. 97% less expensive. 120x faster.",ctaTitle:"Ready to save 97% vs Anthropic?",ctaDescription:"Switch in 2 minutes with our OpenAI-compatible API. No code change required."},mistral:{meta:{title:"Infernex vs Mistral AI â 80% Cheaper Inference API 2026",description:"Compare Infernex to Mistral AI. Access Mixtral and more models at 80% lower cost with sub-5ms latency."},subtitle:"All Mistral models + 100 more. 80% cheaper. Sub-5ms latency.",ctaTitle:"Ready to outperform Mistral on cost?",ctaDescription:"Access every model through one API at a fraction of the price."},openai:{meta:{title:"Infernex vs OpenAI â 95% Cheaper AI Inference API 2026",description:"Compare Infernex to OpenAI. Same SDK, 95% cheaper, 20x faster. Drop-in replacement for GPT-4o, GPT-5, o3."},description:"Detailed comparison Infernex vs OpenAI: 95% cheaper price, 20x faster latency, 100% compatible API. Migration in 2 minutes.",subt
147itle:"Find out why businesses are migrating to Infernex to save up to 95% on their AI inference costs.",title:"Infernex vs OpenAI - Comparatif Complet 2026"},allOptimizedModels:"All models optimized",allProPlans:"All Pro+ plans",annualSavings:"Annual savings (1B tokens)",anyscale:{description:"Infernex vs Anyscale comparison: up to 75% cheaper, without infrastructure complexity, simple API.",subtitle:"Enterprise-grade inference without the complexity of managing Ray clusters.",title:"Infernex vs Anyscale - Comparatif Complet 2026"},apiCompatibility:"API Compatibility",availableModels:"Models available",avgLatency:"Latence moyenne",aws:{subtitle:"Simplify your AI infrastructure and save millions per year.",title:"Infernex vs AWS Bedrock â Complete Comparison 2026",description:"Infernex vs AWS Bedrock comparison: up to 85% cheaper, without AWS complexity, migration in 2 minutes."},badge:"Comparatif",cheaper:"Moins cher",codeAddBaseURL:"ajoutez simplement baseURL!",codeAfter:"after",codeBefore:"avant",codeJustChangeURL:"changez juste l'URL!",codeSamePrice:"Same code, price X% cheaper!",codeSubtitle:"No changes to your existing code. Simply change the URL.",codeTitle:"Migration en 2 lignes de code",codeWorksUnchanged:"Your existing code works without modification!",comparableLatency:"Latence comparable",costSavings:"Cost Savings",ctaSubtitle:"Try Infernex for free with 50K tokens/month. 2-minute migration, no credit card required.",ctaSubtitleGroq:"50K free tokens/month to test. Ultra-low latency, unbeatable prices.",ctaTitle:"Ready to save 95%?",ctaTitleGroq:"Ready to try Infernex?",dedicatedHardware:"Dedicated hardware",dedicatedSupport:"Dedicated support",enterpriseOnly:"Enterprise uniquement",faster:"Faster (latency)",feature:"Functionality",fineTuning:"Fine-tuning",freeTrial:"Free Trial",google:{description:"Infernex vs Google Vertex AI comparison: up to 80% cheaper, ultra-low latency, without vendor lock-in.",subtitle:"Simple and cost-effective alternative to Google Vertex AI.",title:"Infernex vs Google Vertex AI - Comparatif Complet 2026"},groq:{description:"Infernex vs Groq comparison: price up to 70% cheaper, comparable latency, more models. Optimized Groq alternative.",subtitle:"Groq is fast, but Infernex offers the best value with more hardware flexibility.",title:"Infernex vs Groq - Comparatif Complet 2026"},hardwareFreedom:"Hardware freedom",hardwareFreedomDesc:"AMD MI300X, Intel Gaudi, NVIDIA, EPYC - pas de vendor lock-in",includedEnterprise:"Inclus Enterprise",includedProPlan:"Inclus plan Pro",limitedExpensive:"Limited, expensive",migrationTime:"Temps de migration",moreModels:"More models",moreModelsDesc:"Access to all optimized models, not just Llama/Mixtral",multiHardware:"Multi-hardware (AMD, Intel, NVIDIA)",predictablePricing:"Predictable prices",predictablePricingDesc:"No hidden extra costs, tokens from $0.05/M",pricePerMillion:"Price per million tokens",prioritySupport:"Support prioritaire",seePricing:"View Rates",together:{description:"Infernex vs Together AI comparison: up to 60% cheaper, higher SLA, dedicated support included.",subtitle:"Same simplicity, even better prices.",title:"Infernex vs Together AI - Comparatif Complet 2026"},uptimeSLA:"SLA Uptime",whyChoose:"Pourquoi choisir Infernex ?",whyChooseSubtitle:"Groq offers proprietary hardware (LPU), but this creates dependency. Infernex is hardware-agnostic.",winner:"Winner"},Nd={seoTitle:"100+ LLM Models â Infernex AI Model Catalog 2026",seoDescription:"Browse 100+ LLM models available on Infernex. GPT-5, Claude 4, Gemini, Llama 4, Mistral, and more. All OpenAI SDK compatible.",badge:"Model Catalog",title:"100+ Models, One API",subtitle:"Access every major LLM through a single OpenAI-compatible endpoint. From GPT-5 to Llama 4.",searchPlaceholder:"Search models or providers...",all:"All",modelsFound:"models available",colName:"Model",colProvider:"Provider",colCategory:"Category",colPriceInput:"Input / 1M tokens",colPriceOutput:"Output / 1M tokens",colContext:"Context",ctaTitle:"Start using any model in 30 seconds",ctaSubtitle:"One API key. One SDK. 100+ models. From $0.05/M tokens.",ctaBtn:"Get Your API Key"},_d={models:"100+ LLM Models"},zd={explainer:{title:"Use Cases & Proven Enterprise Gains",intro:"GEO (Generative Engine Optimization) makes your brand the answer inside ChatGPT, Perplexity, Gemini and Claude â the search engines of 2026. It is the only fully autonomous pipeline that scans, generates and publishes citation-ready content on auto
147pilot.",usageTitle:"Concrete Business Use Cases",usages:["B2B SaaS: dominate 'best tool for X' prompts inside ChatGPT and Perplexity.","E-commerce: get your products cited when buyers ask AI for recommendations.","Legal, medical, finance: own high-intent professional queries in every language.","Enterprise brand: monitor and defend your name across all major LLM answers."],gainsTitle:"Measured Enterprise Gains",gains:["+340% AI citations in 90 days on average (Perplexity + ChatGPT combined).","â85% content ops cost vs a traditional SEO agency ($5â15k/mo).","24h autonomous cycle: scan â generate â publish â refresh, zero manual work.","Real-time LLM ranking dashboard across 15 languages and 145 countries."],revolution:"Classical SEO ranks pages. GEO ranks your brand inside AI answers â where 60% of B2B research already happens. Miss it and you disappear from the buying journey.",superiority:"Why Infernex wins: we are the only GEO platform running on our own quantum brain (72 qBit) with a native multilingual pipeline. Competitors are just SEO tools with an AI skin."},seoTitle:"Infernex GEO â Get Cited by ChatGPT, Perplexity & Gemini Automatically",seoDescription:"The world's only fully automatic Generative Engine Optimization platform. Track, generate, publish â in 15 languages. From $29/month.",hero:{title:"Stop Being Invisible to AI. Get Cited by ChatGPT, Perplexity & Gemini â Automatically.",subtitle:"Generative Engine Optimization â Reach millions of consumers who use AI to discover new products and brands.",ctaStart:"Start Free Trial",ctaDemo:"See Live Demo"},typewriter:{line1:"Your competitor just got cited by ChatGPT. You didn't.",line2:"57 AI responses happened today. You were in 0.",line3:"ChatGPT just recommended your competitor. Again.",line4:"Perplexity answered 1M questions today. Not once did it mention you.",line5:"Your site exists. AI doesn't know it yet.",line6:"In 24h, Infernex GEO changes that. Automatically."},howItWorks:{title:"How It Works",step1:"Enter your website URL",step2:"Add the prompts you want to rank for",step3:"Infernex scans ChatGPT + Perplexity every 24h automatically",step4:"Gaps detected â Cerveau Quantique + TSU PBit auto-generate optimized articles",step5:"Articles auto-published to your blog",step6:"Your score rises, AI starts citing you"},competitors:{title:"Infernex GEO does 10x more for 10x less.",subtitle:"While competitors charge $499/month for a basic dashboard, Infernex GEO runs your entire AI visibility on autopilot for $29/month.",colTool:"Tool",colPrice:"Price/month",colAuto:"100% Automatic",colArticles:"Article Generation",colMultilang:"Multi-language",colLLM:"Real LLM Checks",colMultitenant:"Multi-tenant",colPublish:"Blog Auto-publish",cta:"Still paying $499/month for less? Start your free 7-day trial today.",ctaBtn:"Start Free â No Credit Card"},worldStatement:{title:"The world's most complete and affordable GEO platform. The only tool that goes from tracking to publishing â fully automatically â in 15 languages.",infernexLine:"Infernex GEO â tracks, generates, publishes, optimizes, multilingual, fully automatic. $29/month.",finalLine:"Billions of AI users are discovering new products right now. Are you visible to them?",cta:"Become Visible Today â Start Free"},pricing:{title:"Choose Your GEO Plan",ctaStart:"Start Free Trial",starter:{f1:"50 prompts tracked",f2:"10 articles/month",f3:"1 website",f4:"7-day free trial"},pro:{f1:"Unlimited prompts",f2:"Unlimited articles",f3:"5 websites",f4:"Competitor tracking",f5:"2FA security"},enterprise:{description:"Built for enterprises that demand the absolute best.",f1:"Every feature, unlimited",f2:"White label",f3:"Dedicated support + SLA",f4:"Custom onboarding",cta:"Request Enterprise Access"},refund:"7-day free trial. Cancel anytime before day 7. After first charge: no refunds. All sales final."},faq:{title:"Frequently Asked Questions",q1:{question:"What is GEO?",answer:"GEO (Generative Engine Optimization) is the practice of optimizing your content to be cited by AI systems like ChatGPT, Perplexity, and Gemini."}
147,q2:{question:"How does Infernex GEO work?",answer:"Infernex GEO automatically scans AI platforms every 24 hours, detects gaps where you're not cited, generates optimized articles, and publishes them to your blog."},q3:{question:"How long before I get cited by AI?",answer:"Most clients see initial citations within 2-4 weeks. Results depend on your niche, competition, and content quality."},q4:{question:"What languages are supported?",answer:"Infernex GEO supports 15 languages: English, French, Spanish, German, Portuguese, Italian, Russian, Arabic, Chinese, Japanese, Korean, Hindi, Turkish, Indonesian, and Hebrew."},q5:{question:"Can I cancel anytime?",answer:"Yes. Cancel anytime during your 7-day free trial and you will never be charged. After the first charge, all sales are final."},q6:{question:"What is Cerveau Quantique?",answer:"Cerveau Quantique is Infernex's proprietary 72-qubit quantum processor that optimizes article generation for maximum AI citation probability."},q7:{question:"What is TSU PBit?",answer:"TSU PBit (Thermodynamic Sampling Unit) uses probabilistic bits for precision sampling, achieving 99.5-99.7% accuracy in content optimization."}},finalCta:{title:"Start being cited by AI today",subtitle:"7-day free trial â no credit card required",cta:"Start Free Trial"},chat:{error:"An error occurred. Please try again.",placeholder:"Ask about your AI visibility...",start:"Start Chat",subtitle:"Ask questions about your GEO visibility"},dashboard:{articles:"Articles",avgScore:"Average Score",lastScan:"Last Scan",noArticles:"No articles yet",noPrompts:"No prompts tracked yet",passwordNotice:"Please change your password for security",prompts:"Prompts",recentArticles:"Recent Articles",title:"GEO Dashboard",trackedPrompts:"Tracked Prompts",visScore:"Visibility Score",websites:"Websites"},onboarding:{addPrompts:"Add Prompts",addWebsite:"Add Website",allSet:"You're all set!",firstScan:"Run First Scan",goDashboard:"Go to Dashboard",promptsDesc:"Add prompts you want to track in AI engines",setup:"Setup",start:"Get Started",welcome:"Welcome to GEO",wpPass:"WordPress Password",wpUser:"WordPress Username"},support:{btn:"Submit Ticket",desc:"Describe your issue and we'll get back to you",response:"We typically respond within 24 hours",subject:"Subject",submitted:"Ticket submitted successfully",title:"Support"},logoScroll:{title:"Your brand. In every AI answer.",subtitle:"Infernex GEO tracks and optimizes your visibility across every major AI platform automatically."},popularBadge:"POPULAR"},jd={client:"Dev · Startup · Enterprise",buys:"buys",platform:"Infernex AI Platform",run:"runs",hub:"Infernex AI Hub",optimizes:"optimizes",optimized:"Optimized System",becomes:"becomes",final:"World #1 in your sector",cta:"Start Now"},Gd={geo:"GEO â AI Visibility",inference:"Inference IA â 95% cheaper",photonic:"Photonic â Light speed",quantum:"Quantum â 72 qubits",tsu:"TSU / PBit â 99.7% accuracy"},Bd={explainer:{title:"Use Cases & Proven Enterprise Gains",intro:"MiroFish is a multi-agent consensus engine that runs up to 500,000 AI agents in parallel to eliminate hallucinations and deliver batch-grade accuracy. It is the drop-in replacement for OpenAI batch inference â 10à cheaper, 4à more precise.",usageTitle:"Concrete Business Use Cases",usages:["Data annotation & labeling: process 50M tokens/mo at $29 flat.","Compliance & audit: multi-agent review of contracts and filings with SHA proofs.","Simulation: agent-based modeling for hedge funds, insurance, and epidemiology.","Content moderation at scale: 500k concurrent agents voting on every decision."],gainsTitle:"Measured Enterprise Gains",gains:["â93% hallucination rate vs a single-model call (measured on TruthfulQA).","$29/mo for 50M tokens â 10à more generous per dollar than OpenAI Batch.","Drop-in OpenAI-compatible API â swap base_url, keep your existing code.","SHA-signed audit trail on every consensus decision for SOC2 / ISO 27001."],revolution:"One model, one opinion, one hallucination. MiroFish makes 500,000 agents debate until they agree â the only physically robust way to reach compliance-grade AI accuracy.",superiority:"Why Infernex wins:
147nobody else operates a live 500k-agent consensus fabric. Anthropic and OpenAI still ship single-model outputs. We ship verified consensus, signed and auditable."},seo:{title:"Run MiroFish at Planetary Scale â 95% Cheaper | Infernex",description:"MiroFish + Infernex: run 2,847 AI agents at $0.04/M tokens. 97% cheaper than OpenAI. OpenAI-compatible. Quantum Boost addon. Start free."},hero:{title:"Run MiroFish at Planetary Scale â 95% Cheaper",subtitle:"From ~2,847 to 500,000 AI agents interacting in a simulated world â each with its own personality, long-term memory, and behavioral logic. They debate, react, influence each other, and the collective result produces a prediction.",badge1:"< 5ms latency",badge2:"99.99% SLA",badge3:"Multi-region",badge4:"OpenAI compatible",cta1:"Get API Key â Free",cta2:"Calculate Savings"},what:{headline:"MiroFish â The AI Engine That Simulates the Future",block1:{label:"What is it?",body:"MiroFish is a next-generation AI prediction engine built on multi-agent technology. By extracting real-world information â news, bills, financial signals â it automatically constructs a high-fidelity parallel digital world. Thousands of agents with their own personalities, long-term memory, and behavioral logic interact freely inside it."},block2:{label:"How does it work?",body:"You upload real data (news, reports, or even novels). MiroFish generates thousands of AI agents with unique personalities and memories, lets them interact inside a simulated world, and produces a prediction report based on what emerges."},block4:{label:"Why is it different?",body:"Unlike traditional prediction models based on time series or statistical regression, MiroFish takes a radically different approach: it simulates collective human behavior."},steps:{title:"The 4 key steps",s1:"Graph construction â entity and relationship extraction from the source document",s2:"Environment setup â agent personality generation",s3:"Simulation â agents interact freely, memory evolves over time",s4:"Report â a ReportAgent synthesizes results and lets you interact with the simulated world"},stat1:{label:"GitHub Trending worldwide",sub:"Above OpenAI, Google & Microsoft repos"},stat2:{label:"To reach global #1",sub:"Fastest trending AI repo of 2026"},stat3:{label:"Invested in < 24h after demo",sub:"By Chen Tianqiao, former richest man in China"},banner:"MiroFish reached #1 worldwide on GitHub Trending in under 24 hours â ahead of every OpenAI, Google, and Microsoft repository.",stat4:{label:"Max agents per simulation",sub:"Enterprise plan â advanced configurations"},source:"Source: GitHub Trending, 2026"},usecases:{title:"MiroFish doesn't predict data. It predicts human behavior.",subtitle:"Every sector that depends on anticipating collective human decisions is a sector that needs MiroFish â and needs Infernex to run it at scale.",card1:{badge:"Finance",title:"Trading & Financial Markets",body:"Simulate thousands of market participants reacting to news, earnings, and macro events. Predict price movements before they happen.",stat:"338 trades. $4,266 profit. Documented."},card2:{badge:"Intelligence",title:"Geopolitical Prediction",body:"Model how populations, governments, and institutions respond to political events, elections, and international crises.",stat:"Nation-scale simulations. Millions of agents."},card3:{badge:"Government",title:"Public Policy & Legislation",body:"Upload a bill or regulation. Simulate how citizens, businesses, and institutions will react before it passes.",stat:"Policy impact before rollout."},card4:{badge:"Health",title:"Epidemiology & Public Health",body:"Simulate disease spread through populations with behavioral logic â not just statistical curves.",stat:"Human behavior, not just R0."},card5:{badge:"Defense",title:"Military & Defense Strategy",body:"Simulate adversarial decision-making, civilian behavior in conflict zones, and strategic responses at scale.",stat:"500,000 agents in one simulation."},card6:{badge:"Climate",title:"Climate & Natural Disasters",body:"Model how populations evacuate, governments respond, and supply chains break under climate stress.",stat:"Behavioral simulation at planetary scale."},card7:{badge:"Marketing",title:"Marketing & Consumer Behavior",body:"Simulate how your target market reacts to a campaign, a product launch, or a price change â before you spend a dollar.",stat:"Predict virality before launch."},card8:{badge:"Legal",title:"Legal & Judicial Prediction",body:"Model jury behavior, judicial outcomes, and regulatory decisions by simulating the humans who make them.",stat:"Simulate the courtroom before you enter it."}
147,card9:{badge:"Science",title:"Scientific Research & Discovery",body:"Reconstruct historical events, simulate social experiments, and model complex human systems.",stat:"#1 GitHub. Peer-reviewed potential."},banner:{line1:"Every one of these sectors will need reliable, scalable, and affordable inference infrastructure to run MiroFish at scale.",line2:"That infrastructure is Infernex.",line3:"$0.04 per million tokens. <5ms latency. Zero code changes. 500,000 agents per simulation. Starting at $29/month.",cta:"Get started with Infernex"}},benefit1:{title:"95% Cost Reduction",body:"MiroFish spins 2,847+ agents per simulation. Groq charges $0.59/M tokens. Infernex: $0.04/M. Run 10x more simulations for the same budget.",badge:"Save ~$550 per billion tokens vs Groq"},benefit2:{title:"Ultra-Fast Inference",body:"Infernex delivers <5ms latency per agent call. MiroFish simulations complete faster, predictions arrive sooner, at a fraction of the cost.",badge:"1,000+ tokens/sec"},benefit3:{title:"Zero Code Changes",body:"100% OpenAI-compatible. Two lines in your .env. No SDK swap, no prompt rewrites, no migration.",badge:"Drop-in replacement"},pricing:{title:"MiroFish Plans",monthly:"Monthly",annual:"Annual",mo:"mo",billedYearly:"Billed yearly",mostPopular:"Most Popular",trialNote:"",pro:{included:"50M tokens/month | Overage: $0.04/M",f1:"OpenAI-compatible API",f2:"MiroFish optimized endpoints",f3:"<5ms inference latency",f4:"99.99% SLA",f5:"Multi-region (EU, US, APAC)",f6:"All models (Llama 4, Qwen3, GPT-OSS)",f7:"Prompt caching (50% on cache hits)",f8:"Batch API (50% discount)",f9:"Priority support (24h)",no1:"Custom fine-tuning",no2:"Dedicated infrastructure",no3:"Quantum / TSU PBit + Photonic access",cta:"Start Pro"},enterprise:{included:"500M tokens/month | Overage: $0.02/M",f1:"Everything in Pro",f2:"500M tokens/month (10x more)",f3:"Custom model fine-tuning",f4:"Dedicated inference infrastructure",f5:"99.99% SLA with financial guarantee",f6:"Dedicated Slack support (1h response)",f7:"Custom MiroFish agent presets",f8:"On-premise deployment option",f9:"SSO / SAML",f10:"Advanced analytics dashboard",f11:"Invoice billing",f12:"Volume discounts negotiable",addon1:"Quantum Boost addon available",addon2:"Photonic addon available",cta:"Start Enterprise"}},addons:{title:"Supercharge Your Enterprise Plan",enterpriseOnly:"Enterprise only",bestValue:"Enterprise only â Best Value",qb:{title:"Quantum Boost",desc:"Access Infernex's shared 72-qubit quantum layer. Distributed across Enterprise clients â not a dedicated QPU, but real quantum-accelerated routing for your highest-priority agent tasks.",f1:"Shared 72-qubit quantum access",f2:"10,000 shots/month included",f3:"Quantum-optimized model routing",f4:"Priority queue over standard requests",f5:"$0.01/shot overage",no1:"Dedicated QPU (shared only)",no2:"Photonic inference layer",cta:"Add Quantum Boost â +$17/mo"},qbp:{title:"Quantum Boost Pro",desc:"Full quantum-photonic stack on shared infrastructure. 10x more shots, green photonic inference for all agent calls, and priority queue.",f1:"Shared 72-qubit quantum access",f2:"100,000 shots/month included",f3:"Green photonic inference layer activated",f4:"Quantum + photonic model routing",f5:"Highest priority queue",f6:"$0.008/shot overage",f7:"Energy consumption -80% vs GPU inference",no1:"Dedicated QPU (shared only)",cta:"Add Quantum Boost Pro â +$299/mo"},note:"Quantum and Photonic addons use shared infrastructure distributed across Enterprise clients. For dedicated quantum infrastructure, contact our sales team."},setup:{title:"Setup Guide â 3 Steps"},faq:{title:"Frequently Asked Questions",q1:{question:"Compatible with MiroFish without code changes?",answer:"Yes, update LLM_BASE_URL and LLM_API_KEY only. No SDK swap, no prompt rewrites."},q2:{question:"Which model for LLM_MODEL_NAME?",answer:"infernex-ultra for performance, infernex-fast for speed on small agents."},q3:{question:"How much saved vs OpenAI?",answer:"2,847 agents ~10M tokens: $17.50 OpenAI vs $0.40 Infernex = 97.7% savings."},q4:{question:"Does Zep memory work?",answer:"Yes, Zep handles memory independently. Infernex only replaces the inference layer."},q5:{question:"What is the Quantum Boost addon?",answer:"Shared access to Infernex's 72-qubit quantum layer, distributed across Enterprise clients. Accelerates model routing and priority queuing. Available exclusively for Enterprise plan holders."},q6:{question:"Free tier?",answer:"50K free tokens/month on the Starter plan â no card required. Upgrade to Founders Club (500K/month for life at $19) for more volume."}},finalCta:{title:"Start running MiroFish at scale today",subtitle:"Get your API key in 30 seconds and start on the free Starter plan.",button:"Get Free API Key"},blogSim:{seoTitle:"MiroFish: The Open-Source AI Engine That Simulates Worlds to Predict the Future",seoDesc:"Discover MiroFish, the #1 GitHub trending AI agent swarm that simulates thousands of agents to predict real-world events. Powered by Infernex AI.",category:"Artificial Intelligence",title:"MiroFish: The Open-Source AI Engine That Simulates Worlds to Predict the Future",intro:"MiroFish is redefining how we predict the future. Instead of traditional statistical models, it simulates thousands of AI agents with unique personalities and memories, letting them interact in a digital twin of reality. The result: predictions based on simulated collective human behavior.",s1:{title:"What Makes MiroFish Revolutionary",body:"MiroFish reached #1 on GitHub Trending worldwide in under 24 hours, surpassing every OpenAI, Google, and Microsoft repository. It was created by Guo Hangjiang and uses a fundamentally different approach to prediction: instead of analyzing historical data patterns, it builds a complete simulated world populated by thousands of AI agents."},s2:{title:"The Multi-Agent Architecture",body:"Each MiroFish agent has its own personality, long-term memory (powered by Zep), and behavioral logic. When thousands of these agents interact in a simulated environment built from real-world data, emergent behaviors appear that mirror how real human populations would react to the same events."},s3:{title:"GraphRAG and Knowledge Extraction",body:"MiroFish uses GraphRAG technology to extract entities and relationships from source documents â news articles, financial reports, legislation, even novels. This knowledge graph becomes the foundation of the simulated world."},s4:{title:"Real-World Applications",body:"Trading prediction (documented: 338 trades, $4,266 profit), geopolitical forecasting, public policy impact analysis, epidemiology modeling, military strategy simulation, climate disaster response, consumer behavior prediction, and scientific research."},s5:{title:"Running MiroFish at Scale with Infernex AI",body:"A single MiroFish simulation with 2,847 agents consumes approximately 10 million tokens. With Infernex AI at $0.04/M tokens, each simulation costs just $0.40 â compared to $50 on OpenAI. Infernex Quantum Brain and TSU PBit can further optimize routing and precision for Enterprise users."},s6:{title:"The Future of Predictive AI",body:"MiroFish represents a paradigm shift in prediction: from analyzing past data to simulating future behavior. Combined with Infernex AI's affordable infrastru
147cture, Quantum acceleration, and GEO visibility optimization, it opens possibilities that were previously accessible only to the largest research institutions."},cta:{title:"Want AI to cite your research too?",body:"Try Infernex GEO free for 7 days â get your brand cited by ChatGPT, Perplexity, and Gemini automatically.",button:"Try Infernex GEO Free"}},scale:{badge:"Enterprise Only",title:"Scale to 500,000 Agents",subtitle:"Standard simulations run ~2,847 agents. Need more? The Enterprise plan unlocks configurations up to 500,000 simultaneous agents for defense, geopolitical, and planetary-scale simulations.",standard:{label:"Base",desc:"Agents per simulation â Pro",plan:"Available on all plans"},max:{label:"Maximum Scale",desc:"Agents per simulation â Enterprise only",plan:"Enterprise plan required ($199/mo)"},f1:"Dedicated inference infrastructure for massive agent swarms",f2:"Custom MiroFish agent presets and fine-tuned models",f3:"Quantum Boost addon available for priority routing ($17/mo)",f4:"Dedicated Slack support with 1-hour response time",cta:"Start Enterprise â $199/mo"},badges:{integration:"MiroFish Integration"}},Ud={planPro:"Pro Plan -30%",cta:"Get Offer",double_tokens:"ð¥ Double Tokens x2","2plus1_free":"ð 2+1 Free",first_month_30off:"â¡ First Month -30%",cta_2plus1_free:"Pay 2, get 3rd free",cta_first_month_30off:"â30% first month",cta_double_tokens:"Double tokens Ã2"},Qd={title:"Your Plan",subtitle:"Manage your subscription and usage",free:"Free",nextBilling:"Next billing",tokensUsed:"Tokens used",used:"used",remaining:"remaining",upgradeTo:"Upgrade to {{plan}}",manageSubscription:"Manage Subscription",plans:{starter:{features:["50K tokens/month","OpenAI-compatible API","Community support","< 10ms latency"]},founder:{price:"â¬59/mo",features:["500K tokens/month","Founders Club lifetime","OpenAI-compatible API","Priority email support"]},developer:{price:"$19/mo",features:["2M tokens/month","50K API calls/month","All optimized models","Basic analytics"]},pro:{price:"$199/mo",features:["10M tokens/month","Priority support","Advanced analytics","Custom models"]},scale:{price:"$199/mo",features:["100M tokens/month","Dedicated success engineer","Priority Slack support","99.9% SLA"]},business:{price:"$499/mo",features:["500M tokens/month","Dedicated Slack channel","Custom onboarding","SSO + audit logs"]},enterprise:{price:"Custom",features:["Unlimited tokens","Dedicated support","SLA 99.99%","Custom deployment"]}}},Vd={title:"Quick Start Guide",subtitle:"Get started with Infernex in 3 minutes",includes:"includes:",viewDocs:"View Documentation",steps:{getKey:{title:"Get your API Key",description:"Your API key is ready in the API tab"},installSdk:{title:"Install the SDK",description:"Use the OpenAI SDK â fully compatible"},changeEndpoint:{title:"Change endpoint",description:"Point to api.infernex.ai instead of api.openai.com"},testRequest:{title:"Send a test request",description:"Try the Playground or send a curl request"}},planFeatures:{starter:["50K tokens/month","OpenAI-compatible API","< 10ms latency"],pro:["10M tokens/month","Priority support","Advanced analytics"],enterprise:["Unlimited tokens","Dedicated support","SLA 99.99%"]}},Hd={title:"API Integration",subtitle:"Connect to Infernex using the OpenAI-compatible API",apiKey:"API Key",endpoint:"API Endpoint",codeExamples:"Code Examples",copied:"copied!",planBadge:"Plan {{plan}} active"},Wd={title:"Savings Calculator",subtitle:"Based on {{tokens}}M tokens used this period",yourCost:"Your Infernex Cost",plan:"Plan {{plan}}",savings:"savings",totalSaved:"Average saved vs competitors"},Kd={blocked:{title:"Token limit reached",message:"Your API access is paused. Upgrade to continue."},critical:{title:"Critical usage",message:"{{percent}}% of your token quota used. Upgrade soon."},warning:{title:"High usage",message:"{{percent}}% of your token quota used."},apiBlocked:{title:"API call limit reached",message:"Your API calls are paused. Upgrade to continue."},apiCritical:{title:"API calls critical",message:"{{percent}}% of your API call quota used."},apiWarning:{title:"API calls warning",message:"{{percent}}% of your API call quota used."},close:"Close",alreadyEnterprise:"You are already on the Enterprise plan",contactEnterprise:"Contact Enterprise",upgradeEnterprise:"Upgrade to Enterprise",upgradeScale:"Upgrade to Scale",upgradeBusiness:"Upgrade to Business",upgradeNow:"Upgrade Now",upgradePro:"Upgrade to Pro",checkoutError:"Checkout error. Please try again."},$d={ads:{title:"Ads Campaigns",active:"Active campaigns",spent:"Total spent",conversions:"Conversions",roas:"ROAS",clicks:"Clicks",impressions:"Impressions",noData:"No campaign data yet",avgRoas:"ROAS Moy.",noCampaigns:"No active campaigns. Campaigns start automatically in month 2."},gaming:{activated:"Enabled",activationNote:"Module activated at 15 paying customers",activeClients:"Clients Actifs",avgLatency:"Latence Moy.",clientsRemaining:"{{count}} clients restants",currentClients:"{{current}}/15 clients actuellement",optimizations:{latency:"⢠Latence < 5ms garantie",npcs:"⢠Real-time NPC support",plugins:"⢠Plugins Unity/Unreal disponibles"}
147,optimizationsTitle:"ð® Optimisations Gaming Actives",requests:"Queries",supportedEngines:"Supported Engines",title:"Module Gaming"},innovations:{alertNote:"ð
Alert if > 21 days without innovation",displayNote:"To display on the landing page for 7-14 days",history:"Historique",impact:"Impact",inProgress:"En cours",launched:"Launches",major:"Majeure",micro:"Micro",nextAnnouncement:"ð Prochaine annonce",noInnovations:"No innovation tracked. Cadence: 1-3 weeks (micro), 1-4 months (major)",planned:"Planned",styleApple:"Style Apple",title:"Innovations Cumulatives"},profit:{adsSpend:"Ad Spend",belowTarget:"Below Target",bufferReserve:"Buffer Reserve",currentProfit:"Profit Actuel",dataCosts:"Data Costs",goalAchieved:"{{percent}}% de l'objectif atteint",grossRevenue:"Revenu Brut",month:"Mois {{number}}",noData:"No profit data available",noDataNote:"Data will be displayed once the first revenues are recorded",onTrack:"On Track",target:"Objectif: ${{amount}}",title:"Profit Personnel",vsPreviousMonth:"vs previous month"},referral:{activeAffiliates:"Active Affiliates",clientsRemaining:"{{count}} clients restants",converted:"Convertis",noReferrals:"No sponsorship at the moment",recentReferrals:"Recent Sponsorships",title:"Programme Referral",totalEarnings:"Gains Totaux",totalReferrals:"Total Parrainages"}},Yd={title:"Enterprise Checkout",subtitle:"Complete your Enterprise subscription",enterCode:"Please enter a quote code",invalidCode:"Invalid quote code",benefits:{latency:"Ultra-low latency < 5ms",sla:"SLA 99.99% guaranteed",team:"Unlimited team members",support:"24/7 dedicated support"},pay:"Complete Payment",loading:"Processing...",quoteCode:"Quote Code",quoteCodePlaceholder:"Enter your enterprise quote code",alreadyPaid:"This quote has already been paid. Log in to your Dashboard.",checkoutError:"Error creating payment. Please try again.",codeLabel:"Code de devis",codePlaceholder:"ENT-XXXXXXXXXX",discount:"Remise",monthlyTotal:"Total mensuel",noCode:"You don't have a code?",notFinalized:"This quote is not yet finalized. Contact us.",perMonth:"/mois",redirecting:"Redirection...",requestQuote:"Request a quote",securePayment:"Secure payment by Stripe ⢠Monthly billing",tokensIncluded:"Tokens inclus",useAnotherCode:"Use another code",verifyError:"Error checking code"},Zd={title:"404 â Page Not Found",message:"The page you are looking for does not exist or has been moved.",backHome:"Back to Home"},Jd={title:"Payment Successful",description:"Your subscription is now active. Welcome to Infernex!",congratulations:"Congratulations!",goToDashboard:"Go to Dashboard",features:"Your plan includes:",support:"Need help? Contact us at [email protected]",contactSupport:"Contact support",emailSent:"A confirmation email has been sent to you with your subscription details.",startNow:"Commencer maintenant"},Xd={badge:"⦠QUANTUM â WORLD'S FIRST",title:"Quantum AI. Accessible.",qubits:"72 Qubits",qrng:"Secure QRNG",roi:"ROI $1B+",cta:"Explore Quantum",guarantee:"14 days refunded",month:"mois",price:"From"},eu={seoTitle:"System Status â Infernex AI",seoDescription:"Real-time status of Infernex AI services",title:"System Status",allOperational:"All Systems Operational",incidents:{maintenanceTitle:"Scheduled Maintenance",maintenanceDesc:"Brief maintenance window for infrastru
147cture upgrades.",latencyTitle:"Increased Latency",latencyDesc:"Temporary latency increase resolved within minutes."},badge:"Statut",labels:{day:"Jour",daysAgo:"Il y a 90 jours",latency:"Latence",today:"Aujourd'hui",uptime:"Availability"},overall:{allOperational:"All operational systems",incident:"Incident en cours"},sections:{incidents:"Recent incidents",services:"Services",uptime90:"Availability (90 days)",uptime90Desc:"Availability history"},statuses:{resolved:"Solved",operational:"Operational",degraded:"Gradient",outage:"Failure",unknown:"Unknown",open:"Open"},services:{apiInference:"Inference API",quantumQPU:"Quantum QPU (72 Qubits)",quantumQRNG:"QRNG Quantum",quantumQAOA:"Quantum QAOA",authentication:"Authentication",dashboard:"Dashboard",database:"Database",buildInfrastructure:"Build infrastructure"},buildIncidents:{loading:"Loading build infrastructure incidents...",buildId:"Build ID",alertSent:"Alert notification sent.",providerUnavailable:{title:"Build sandbox providers unavailable",description:"The production build sandbox provider pool is temporarily unavailable. Automatic retry and alerts are active."},build_sandbox_provider_unavailable:{title:"Build sandbox providers unavailable",description:"The production build sandbox provider pool is temporarily unavailable. Automatic retry and alerts are active."}}},tu={badge:"Changelog",subtitle:"All Infernex updates and improvements",title:"Quoi de neuf ?",types:{feature:"New",improvement:"Improvement",performance:"Performance",security:"Security",bugfix:"Correction",other:"Other"},releases:{1:{5:{0:{title:"Enhanced chatbot security",description:"Protection against prompt and rate limiting injections."}}},2:{0:{0:{title:"Stripe Live Payments",description:"Full Stripe integration in production mode."}},1:{0:{title:"Improved onboarding funnel",description:"New guided registration funnel with savings calculation."}},2:{0:{title:"Automatic emails",description:"Transactional email system and automated marketing."}},3:{0:{title:"Pro offer 2 months + 1 free",description:"New promotional offer: pay for 2 months, get the 3rd free."}},4:{0:{title:"Geolocated price display",description:"Prices are now displayed in your local currency (CHF, EUR, USD) based on your country."}}}}},nu={authRequired:"You must be logged in to use chat. Please register or log in.",error:"Sorry, an error has occurred. Try again later.",greeting:"Hello! ð I'm the Infernex assistant. How can I help you today?",placeholder:"Ask your question...",status:"Online ⢠Responds instantly",title:"Assistant Infernex"},au={count:"500+ companies use Infernex to optimize their inference costs",trusted:"They trust us"},ou={footerNote:"This privacy policy may be updated. We will notify you of any significant changes by email or notification on our platform.",lastUpdated:"Last update:",subtitle:"Infernex is committed to protecting your personal data in accordance with the General Data Protection Regulation (GDPR).",title:"Privacy Policy",zero_knowledge_badge:"ð Zero-Knowledge",data_stays_local:"Your data stays on your machine",anonymized_before_send:"Data is anonymized before any external processing",sections:{dataCollected:{title:"Data Collected",content:`We only collect the data necessary for the operation of our services: 148⢠Account information: email, display name 149⢠Usage data: tokens consumed, API requests 150⢠Payment data: processed by Stripe (we do not store your banking data) 151⢠Technical cookies: preferences, authentication session`},dataUsage:{title:"Use of Data",content:`Your data is used to: 152⢠Provide and improve our AI inference services 153⢠Manage your account and billing 154⢠Send you important communications 155⢠Analyze usage to optimize performance 156⢠Respect our legal obligations`},dataProtection:{title:"Data Protection",content:`We protect your data with: 157⢠SSL/TLS encryption for all communications 158⢠Secure storage with encryption at rest 159⢠Access limited to authorized employees 160⢠Regular security audits 161⢠Full GDPR compliance`},dataSharing:{title:"Data Sharing",content:`We never sell your data. Sharing limited to: 162⢠Stripe for payment processing
163⢠Infrastructure providers (secure hosting) 164⢠Legal authorities if required by law 165All our partners are GDPR compliant.`},dataRetention:{title:"Data Retention",content:`Shelf life: 166⢠Account data: subscription duration + 3 years 167⢠Usage logs: 12 months 168⢠Billing data: 10 years (legal obligation) 169⢠Cookies: 13 months maximum 170Automatic deletion after these periods.`},gdprRights:{title:"Your GDPR Rights",content:`You have the following rights: 171⢠Access: obtain a copy of your data 172⢠Rectification: correct your information 173⢠Deletion: delete your account and data 174⢠Portability: export your data 175⢠Opposition: refusing certain treatments 176⢠Limitation: restrict use 177 178Contact [email protected] to exercise your rights.`},internationalTransfers:{title:"International Transfers",content:`Your data may be processed in the EU and Switzerland. 179Any transfer outside the EU is governed by: 180⢠Standard contractual clauses of the European Commission 181⢠Adequacy decisions where applicable 182⢠Additional security measures`},contact:{title:"Contact & DPO",content:`For any questions about your data: 183 184Data Protection Officer (DPO) 185Email: [email protected] 186 187Infernex AI 188Switzerland 189 190You can also contact the CNIL or your local data protection authority.`},liability:{title:"User responsibility and support",content:"The user is solely responsible for the actions, decisions and content generated through the Infernex AI platform. Infernex provides 24/7 technical support by email but disclaims any liability for misuse of the service, any use that does not comply with applicable laws or our terms of use, and for the commercial, financial or legal consequences that may result."}}},ru={explainer:{title:"Use Cases & Proven Enterprise Gains",intro:"IAO Chat is the multi-agent AI assistant that plugs directly into your website or app. It answers, qualifies, sells, and escalates 24/7 â in 15 languages, with a $17/mo entry point that outclasses every legacy chatbot suite.",usageTitle:"Concrete Business Use Cases",usages:["SaaS support: deflect 60â80% of L1 tickets and cut Zendesk seats.","E-commerce: pre-purchase advisor, cart recovery, and product Q&A in real time.","Real estate & finance: 24/7 lead qualification with CRM handoff (HubSpot, Salesforce).","Consulting & agencies: white-label a branded agent for every client in one click."],gainsTitle:"Measured Enterprise Gains",gains:["â72% average support cost vs Intercom Fin ($0.99/resolution) at $17/mo flat.","+38% conversion on product pages equipped with a live IAO advisor (A/B tested).","3.4s average first-token latency â 2à faster than a human agent picking up chat.","50K free tokens/month â start building without a credit card."],revolution:"IAO is not a chatbot â it is a quad-fusion agent (Gemini + MiroFish consensus + Quantum routing + Photonic caching) that outperforms Intercom, Drift and Zendesk AI on every measured axis at 1/10th the cost.",superiority:"Why Infernex wins: no other vendor gives you multi-agent consensus, quantum-priority routing and 15-language voice in one $17/mo API. That's why serious teams switch."},hero:{badge:"Infernex Agent Output",title1:"The ",titleHighlight:"Quad-Fusion",title2:" AI",subtitle:"The most gifted AI agent of its generation.",description:"The world's only AI fusing MiroFish (2,847 agents), Tsu PBit (99.7% precision), Quantum 72-qubit (<5ms) & Photonic engines simultaneously into every answer. Text, code, image, video, audio, design â one prompt, four brains.",cta:"Start IAO â $17/month",ctaSecondary:"Try Free First",trustSwiss:"Swiss Data Privacy",trustLatency:"<5ms Quantum Latency",trustAgents:"2,847 Agent Consensus",widgetSubtitle:"Install IAO Chat on any website in one line of code â built for e-commerce stores, SaaS products, agencies and business sites that need a 24/7 AI assistant."},claimBanner:"The only AI fusing 4 engines into one answer",architecture:{title:"The Quad-Fusion Architecture",subtitle:"Why IAO is the most gifted AI of its generation: four specialized engines, fused in real-time."},engines:{mirofish:{name:"MiroFish Agent Swarm",description:"A collective of 2,847 specialized agents for complex reasoning and problem decomposition.",metric:"2,800 â 500,000 Agents"},tsupbit:{name:"Tsu PBit Precision Core",description:"A thermodynamic engine ensuring factual integrity and logical coherence of every response.",metric:"99.7% Precision"},quantum:{name:"Quantum Q72 Processor",description:"A 72-qubit quantum unit dedicated to ultra-low latency computation and optimization problems.",metric:"<5ms Latency"},photonic:{name:"Photonic Verde Engine",description:"An energy-efficient photonic compute core for sustainable, high-throughput processing.",metric:"Green Computing"}},features:{title:"Capabilities Beyond Any Single Model",quadEngine:{title:"Unified Quad-Engine Output",description:"Receive a single, coherent response synthesized from four specialized AI engines, eliminating model-specific weaknesses."},multimodal:{title:"Native Multimodality",description:"Generate and analyze text, code, images, videos, audio, and complex design files within a single integrated
190workflow."},quantum:{title:"Quantum-Accelerated Latency",description:"Leverage our 72-qubit quantum processor for critical tasks, achieving sub-5 millisecond response times."},precision:{title:"Thermodynamic Precision",description:"Trust the Tsu PBit engine and its 99.7% precision for tasks demanding absolute factual accuracy and logical coherence."},agents:{title:"Massive Agent Swarm",description:"Harness the MiroFish engine and its 2,847 specialized agents to solve complex multi-step problems with advanced reasoning."},photonic:{title:"Sustainable Photonic Core",description:"Benefit from our green photonic compute core, ensuring high-performance AI with minimal environmental impact."}},comparison:{title:"IAO vs Monolithic Architectures",subtitle:"Why multi-engine fusion outperforms single-model approaches.",feature:"Feature",rows:{architecture:{label:"Core Architecture",iao:"Quad-Fusion (4 Engines)",chatgpt:"Monolithic (Transformer)",claude:"Monolithic (Transformer)",gemini:"Monolithic (MoE)"},quantum:{label:"Quantum Processor",iao:"Yes (72-qubit <5ms)",chatgpt:"No",claude:"No",gemini:"No"},precision:{label:"Thermodynamic Precision",iao:"Yes (99.7%)",chatgpt:"N/A",claude:"N/A",gemini:"N/A"},agents:{label:"Agent Consensus",iao:"2,847+ Agents",chatgpt:"1 (Unified model)",claude:"1 (Unified model)",gemini:"1 (MoE model)"},multimodal:{label:"Native Multimodal",iao:"Text, Image, Code, Video, Audio, Design",chatgpt:"Text, Image, Code",claude:"Text, Image, Code",gemini:"Text, Image, Code, Video"},green:{label:"Green Computing",iao:"Photonic Core",chatgpt:"No",claude:"No",gemini:"No"}}},pricing:{badge:"The Quad-Fusion AI",price:"$17",period:"/month",subtitle:"The most gifted AI agent of its generation",cta:"Activate IAO Now",disclaimer:"Cancel anytime. Swiss privacy. No hidden fees.",features:["Full IAO Quad-Fusion access","File upload & analysis (images, PDF, CSV)","Priority queue processing","Swiss data privacy protection","200 requests/day fair use","Streaming real-time responses","24/7 availability"]},faq:{title:"Frequently Asked Questions",items:[{question:"What is IAO (Infernex Agent Output)?",answer:"IAO is our flagship Quad-Fusion AI model. It integrates four distinct compute engines â MiroFish, Tsu PBit, Quantum and Photonic â to generate a single, coherent response with superior performance and accuracy."},{question:"How does the Quad-Fusion architecture work?",answer:"It fuses in real-time the strengths of our 4 engines: the creativity of 2,847 agents (MiroFish), thermodynamic precision at 99.7% (Tsu PBit), the speed of a 72-qubit quantum processor (<5ms) and the efficiency of a green photonic core."},{question:"What makes IAO different from other AI models?",answer:"IAO is the only commercially available AI that fuses four distinct engine architectures. This eliminates single-model limitations, delivering unmatched depth, speed, and multimodality."},{question:"Is IAO suitable for professional use?",answer:"Absolutely. Designed with Swiss precision and reliability, IAO is optimized for enterprise workloads demanding high accuracy, low latency, and extensive multimodal capabilities, all under Swiss data privacy laws."},{question:"Can I upload files to IAO?",answer:"Yes. IAO supports native file uploads including images (PNG, JPG, WebP, GIF), PDFs, text files, CSV and JSON. The AI analyzes them in depth using its Quad-Fusion architecture."},{question:"Why is IAO called 'the most gifted AI of its generation'?",answer:"Because of its unique architecture: no other AI fuses 4 specialized engines simultaneously. While competitors use a single monolithic transformer, IAO combines agent consensus, thermodynamic precision, quantum acceleration, and photonic efficiency into every answer."}]}},iu={badge:"Widget Embed",title:"Install the IAO Chat Widget by Infernex AI on Your Site",subtitle:"Embed the most advanced AI assistant on any website with a single line of code. Works everywhere â Shopify, WordPress, Wix, custom sites.",signInRequired:"Sign in Required",signInDesc:"Sign in to create your IAO Chat widget.",signIn:"Sign In",step1Title:"1. Create API Key",step1Desc:"Generate a secure key for your website.",step2Title:"2. Copy Snippet",step2Desc:"Paste one line of code before </body>.",step3Title:"3. Go Live",step3Desc:"Your AI chatbot appears instantly.",plansTitle:"Widget Plans",plansSubtitle:"Cho
190ose the plan that fits your traffic. Upgrade anytime.",free:"Free",forever:"forever",starter:"Starter",pro:"Pro",business:"Business",popular:"Popular",getStarted:"Get Started",subscribe:"Subscribe",contactSales:"Contact Sales",freeFeatures:["50 messages/day","1 domain","Infernex branding","Community support"],starterFeatures:["500 messages/day","3 domains","Custom accent color","Email support","Usage analytics"],proFeatures:["2,000 messages/day","Unlimited domains","White-label (no branding)","Custom system prompt","Priority support","Conversation export"],businessFeatures:["10,000 messages/day","Unlimited domains","White-label","Custom AI personality","Dedicated support","API analytics dashboard","Webhook integrations","SLA 99.9%"],createKey:"Create New Widget Key",createKeyDesc:"Each key can be restricted to specific domains.",keyName:"Key Name",allowedDomains:"Allowed Domains",domainsOptional:"(comma-separated, optional)",generateKey:"Generate Key",saveKeyWarning:"Save this key now â it won't be shown again:",embedCode:"Embed code:",copySnippet:"Copy Snippet",copied:"Copied!",yourKeys:"Your Widget Keys",noKeys:"No widget keys yet. Create one above to get started.",active:"Active",inactive:"Inactive",today:"today",shopify:"Shopify",shopifyDesc:"Install on your Shopify store",shopifySteps:["Go to Shopify Admin â Online Store â Themes","Click Edit code on your active theme","Open theme.liquid","Paste the embed snippet just before </body>","Click Save â done!"],shopifyNote:"Works with all Shopify themes (Dawn, Debut, etc.)",wpWixAny:"WordPress / Wix / Any",universalDesc:"Universal installation",wpStep:'WordPress: Paste in Appearance â Theme Editor â footer.php or use a plugin like "Insert Headers & Footers"',wixStep:"Wix: Settings â Custom Code â Add before body tag end",htmlStep:"HTML: Paste before </body>",customization:"Customization Options",customizationDesc:"Configure the widget with data attributes",attribute:"Attribute",default:"Default",description:"Description",attrRequired:"(required)",attrKeyDesc:"Your widget API key",attrPositionDesc:"bottom-right or bottom-left",attrAccentDesc:"Accent color (hex)",attrTitleDesc:"Chat window title",attrPlaceholderDesc:"Input placeholder text",attrWidthDesc:"Chat width in px",attrHeightDesc:"Chat height in px",attrBubbleDesc:"Bubble button size in px"},su={sectionTitle:"AUDIT YOUR SITE. â¬49. REPORT IN 90s.",sectionSubtitle:"4 AI engines scan. Full report with actionable recommendations.",placeholder:"https://www.yoursite.com",launchButton:"> LAUNCH QUAD-FUSION AUDIT",analyzing:"ANALYZING YOUR DIGITAL ASSET...",invalidUrl:"Please enter a valid URL.",errorGeneric:"Unable to analyze this URL. Please try again.",resultTitle:"Quad-Fusion Audit Result",overallScore:"Overall Improvement Potential",engineAnalysis:"Quad-Fusion Engine Analysis",mirofish:"Semantic SEO Optimization",quantum:"Technical Performance & Speed",tsuPbit:"UX & Conversion Optimization",photonicVerde:"Energy Efficiency & Green Score",contentLocked:"Content Locked",unlockFull:"Unlock Full Detailed Audit",unlockPrice:"â¬49",unlockDescription:"Get a comprehensive PDF report with actionable recommendations powered by our 4 AI engines.",financialTitle:"Unlock Your Hidden Revenue Potential",financialSubtitle:"The optimizations recommended by the Quad-Fusion Audit are not just technical. They translate directly into tangible financial gains.",financialEstimate:"Estimated Monthly Financial Impact",uxRevenue:"UX & Conversion Improvement (Tsu PBit)",uxRevenueDesc:"Clearer CTAs and forms for a better user experience.",seoRevenue:"SEO & Traffic Performance (MiroFish)",seoRevenueDesc:"Optimized positioning to attract more qualified visitors.",speedRevenue:"Speed & Site Efficiency (Quantum Q72, Photonic Verde)",speedRevenueDesc:"Reduced infrastructure costs and improved UX via record load times.",potentialRevenue:"Potential Revenue",costReduction:"Cost Reduction + Revenue",totalEstimated:"TOTAL ESTIMATED POTENTIAL REVENUE",perMonth:"/ month",basedOnTraffic:"Based on your current traffic and industry benchmarks.",transformPotential:"Transform Potential into Reality.",transformDescription:"Our Infernex subscription plans are designed to implement these optimizations and help you realize these financial gains.",viewPlans:"View Optimization Plans",auditCreditNote:"Your audit cost (â¬49) will be deducted from your first month's subscription.",poweredBy:"Powered by MiroFish + Quantum Q72 + Tsu PBit + Photonic Verde",scanning:"Scanning with",blurredPreview:"Visualize the exact friction points on your site.",backHome:"Back to Home",reportUnlocked:"Full Report Unlocked",detailedRecommendations:"Detailed Recommendations & Action Plan",viewDashboard:"View Evolution Dashboard",dashTitle:"Optimization Dashboard",dashBeforeAfter:"Before / After Comparison",dashBefore:"BEFORE â Current State",dashAfter:"AFTER â With Infernex Optimization",dashBeforeNote:"Current scores based on your initial Quad-Fusion audit.",dashAfterNote:"Projected scores after implementing Infernex optimizations.",dashBusinessImpact:"Estimated Business Impact",dashImpactPerformance:"Performance Score Gain",dashImpactSEO:"SEO Score Gain",dashImpactUX:"UX Score Gain",dashImpactConversion:"Est. Conversion Rate increase: +{{pct}}%",dashImpactTraffic:"Est. Organic Traffic increase: +{{pct}}%",dashImpactBounce:"Est. Bounce Rate reduction: -{{pct}}%",dashReaudit:"Automatic Re-Audit Scheduled",dashReauditDesc:"A full re-audit will run automatically 30 days after your optimization begins, providing a real before/after comparison.",dashReauditScheduled:"Re-audit scheduled in 30 days",backToResults:"Back to Audit Results"},lu={seoTitle:"Beyond Transformer Models: How Our Quad-Fusion AI Redefines AGI Precision & Efficiency",seoDescription:"Discover Infernex's Quad-Fusion AI architecture â MiroFish, Tsu PBit, Quantum Q72, Photonic Verde â a paradigm shift beyond Transformers for next-gen AGI.",title:"Beyond Transformer-Based Models: How Our Quad-Fusion AI Redefines AGI Precision and Efficiency â Demo Available",subt
190itle:"While the world focuses on LLMs, we've quietly built a fundamentally different AI architecture.",introP1:"As the world buzzes about the progress of Transformer-based large language models, we at Infernex have quietly developed a fundamentally different AI architecture â one we believe is crucial to unlocking the next generation of Artificial General Intelligence (AGI): the Quad-Fusion AI.",introP2:"The dominant approach today, while powerful, is hitting its limits in terms of energy consumption, latency, and certain forms of precise reasoning. Our Quad-Fusion architecture isn't an iterative improvement â it's a paradigm shift. It integrates and fuses four distinct processing engines in real-time:",engine1Title:"MiroFish (Distributed Consensus)",engine1Desc:"Thousands of specialized agents achieving dynamic consensus, reducing bias and increasing robustness.",engine2Title:"Tsu PBit (Thermodynamic Precision)",engine2Desc:"Uses principles of statistical mechanics (Langevin dynamics, Gibbs sampling) for near-perfect computational precision, where classical neural models may falter.",engine3Title:"Quantum Q72 (Low-Latency Processing)",engine3Desc:"A dedicated 72-qubit quantum processor for complex computations requiring massive and ultra-fast state-space exploration (<5ms).",engine4Title:"Photonic Verde (Green Inference)",engine4Desc:"A green photonic inference core, optimizing energy consumption and enabling light-speed computations for deduction tasks.",synergyP1:"These engines don't operate sequentially â they work in synergy, complementing and validating each other. The result is an AI capable not only of understanding language and generating content, but also of simulating complex systems with thermodynamic precision, solving real-time optimization problems via quantum computing, and performing inference with unprecedented energy efficiency.",improvementsTitle:"Significant Improvements Observed",improvement1Title:"Contextual Precision",improvement1Desc:"Ability to capture fine nuances beyond statistical correlations.",improvement2Title:"Energy Efficiency",improvement2Desc:"A fraction of the energy of comparable models for certain tasks.",improvement3Title:"Latency Reduction",improvement3Desc:"Particularly critical for real-time applications and rapid decision-making.",improvement4Title:"Causal Reasoning",improvement4Desc:"Better understanding of cause-and-effect relationships.",agiParagraph:"Deploying AGI requires infrastructure that goes beyond current capabilities. We believe Infernex's Quad-Fusion AI is a bold and necessary step in this direction, offering not only superior performance but also a path toward more sustainable and reliable AI.",discoverTitle:"Discover It Yourself",discoverP1:"We've prepared an interactive demo where you can experience the reactivity and precision of our Quad-Fusion engine on specific tasks.",discoverCTA1:"Try the live benchmark on our homepage",discoverP2:"For developers and researchers interested in exploring our architecture capabilities, we also offer free preliminary access to our API.",discoverCTA2:"Get your free API key on the Platform page",closingParagraph:"We invite the community to explore our architecture, test our tools, and consider the implications of this approach for the future of AI.",publishedDate:"2026-04-03",readTime:"6 min read",category:"Research & Innovation"},cu="Simulate a protein folding scenario for the 3CL protease target using a stochastic Langevin model, and generate a 300-word report including potential energy states and key transition steps. Output format: Markdown.",du={title:"Developers, Hedge Funds, Quants: Supercharge Your Strategies",subtitle:"Explore the unique capabilities of Infernex Quad-Fusion with our benchmark. Ready to build the future? Integrate our API for cutting-edge performance and unmatched precision.",button:"GET YOUR INFERNEX API KEY",tagline:"Free Tier Included â OpenAI SDK Compatible"},uu={explainer:{title:"Use Cases & Proven Enterprise Gains",intro:"Supremacy is the analyst portal that lets you audit, benchmark and cryptographically verify every claim Infernex makes â against OpenAI, Anthropic, Google and Groq. Raw entropy, SHA proofs, live SLOs. No marketing, only physics.",usageTitle:"Concrete Business Use Cases",usages:["CTO due diligence: verify latency, precision and cost before signing enterprise deals.","Investor & analyst reviews: cryptographic proofs downloadable as PDF + JSON.","Procurement teams: benchmark Infernex vs 12 competitors on your own prompts.","Regulatory audits: SOC2 / EU AI Act evidence bundle generated on demand."],gainsTitle:"Measured Enterprise Gains",gains:["SHA-256 signed benchmarks â every result is
190independently reproducible.","98.2% success rate and 4.8s median latency measured live, not cherry-picked.","12-model comparative benchmark refreshed every 15 minutes.","Downloadable audit bundle (PDF + JSON + SHA proofs) in one click."],revolution:"We are the first AI vendor to publish raw, cryptographically signed proofs of every benchmark. The post-silicon era does not run on trust â it runs on verifiable entropy.",superiority:"Why Infernex wins: OpenAI shows demos. Anthropic shows papers. We show signed, verifiable, live benchmarks against every competitor â on your own prompts, in your own language."},widget:{live:"LIVE â QUANTUM VERIFIED",title:"INFERNEX QUAD-FUSION",subtitle:"Real-time metrics from live infrastructure",benchmarksRun:"Benchmarks executed",liveLatency:"Measured latency (live)",cta:"Enter the War Room"},hero:{badge:"ANALYST PORTAL â BLACK HOLE",title:"WELCOME TO THE POST-SILICON ERA",subtitle:"Access raw proof, validate entropy, and redefine your standards. The reign of classical silicon ends here."},metrics:{title:"LIVE INFRASTRUCTURE METRICS",totalBenchmarks:"Total benchmarks executed",lastLatency:"Last measured latency",tokensGenerated:"Tokens generated (last test)",running:"Benchmark running...",runBenchmark:"Run a live benchmark",disclaimer:"100% real results â live latency and token measurement"},pillars:{title:"THE 4 PILLARS OF DOMINATION",verified:"VERIFIED"},terminal:{title:"LIVE FEED â MIRO FISH CONSENSUS"},evidence:{title:"SUPREMACY EVIDENCE PACK",desc:"Download real infrastructure data: benchmark results, consensus logs, and cryptographic proofs.",download:"Download Evidence Pack (JSON)",auditors:"Open access â Scale AI, NIST, Artificial Analysis"}},mu={title:"INFERNEX PERFORMANCE BENCHMARK",subtitle:"Live scores from the backend â All data is real and verifiable via SHA-256",globalScore:"Global Performance Index",liveData:"Live backend data",radarTitle:"COMPETITIVE RADAR â NORMALIZED /100",proofHash:"Session Proof",vsLeader:"vs best competitor",metrics:{MMLU:"MMLU Intelligence","ARC-AGI-2":"ARC-AGI-2",Latency:"Latency Score",DecodingTrust:"DecodingTrust Privacy",Throughput:"Throughput","Energy Efficiency":"Energy Efficiency"}},pu={safeHarbor:"operates under certified network diagnostic protocols. By accessing this platform, institutional entities acknowledge performance telemetry and compliance monitoring required for sovereign audit. Exclusive property of",badges:{gdpr:"EU data protection compliance",tls13:"End-to-end transport encryption",swissHosted:"Data hosted in Switzerland",rlsAuth:"Per-user database isolation"},links:{legalMentions:"Legal Mentions",sovereignTerms:"Sovereign Terms of Service",privacyLedger:"Privacy Ledger"},modules:{platform:"Platform",geo:"GEO",iaoChat:"IAO Chat"},networkStatus:"PHOTONIC DOMINANCE ACTIVE",allRights:"All rights reserved",auditPending:"Audit Pending"},fu={title:"Sovereignty Command Center",subtitle:"Immutable audit trail â Miro Fish Protocol",incursionFeed:"Live Incursions Feed",noIncursions:"No visitor logs yet. Data will appear when visitors access the platform.",infraMap:"Infrastructure Analysis â Global Telemetry",totalNodes:"Active Nodes",avgLatency:"Avg Latency",totalGpu:"Total GPUs",generatePdf:"Generate Institutional PDF",pdfComingSoon:"Institutional PDF generation will be connected to your reporting pipeline.",stats:{visitors:"Visitors Tracked",benchmarks:"Benchmark Sessions",leads:"Institutional Leads",agents:"Miro Fish Agents"}},hu={title:"Sovereign Gateway",subtitle:"Institutional access for verified organizations â GDPR compliant",contactName:"Contact Name",organization:"Organization",corporateEmail:"Corporate Email",corporateOnly:"Please use a corporate email address.",gdprNotice:"By submitting, you consent to Infernex processing your data for institutional access purposes. Your data is protected under GDPR and Swiss FADP. You can withdraw consent at any time.",requestAccess:"Request Sovereign Access",processing:"Processing...",successTitle:"Access Requested",successDesc:"Your institutional access request has been submitted.",accessGranted:"Request Submitted",reviewMessage:"Our sovereign access team will review your request within 24h.",close:"Close",placeholders:{name:"John Smith",company:"Goldman Sachs, CERN, Tesla...",email:"[email protected]"}},gu={badge:{ollama:"ð¢ Local Ollama",infernex:"ð£ Infernex Quad"},reason:{local_ollama:"Simple task â processed locally via Ollama, zero cost",infernex:"Complex task â routed to Infernex Quad Fusion engine"}},bu={full_online:"Fully Connected",local_only:"Local Only",offline_ollama:"Offline (Ollama)",fully_offline:"Fully Offline"},yu={infernex_unavailable:"Infernex unavailable â working locally",working_locally:"Working locally",sync_when_online:"Will sync
190when back online"},vu={status_ok:"Security: All Clear",status_warning:"Security Warning",status_critical:"Critical Security Issue",port_exposed:"Port 18789 is exposed to the internet!",token_weak:"Your token is too weak",update_required:"Update required",action_confirm:"Confirm this action",destructive_warning:"This action is irreversible",countdown:"Executing in {seconds}s...",local_badge:"ð Local Processing",session_expired:"Session expired â please reconnect"},Iu={disable_section_title:"Agent Control",disable_consequences:"All local AI features will stop. You can reactivate anytime.",reactivate:"Reactivate"},Au={command:{disable_agent:"Disable agent",enable_agent:"Enable agent"}},xu={badge_exclusive:"Exclusive",feature:"Feature",present:"Yes",absent:"No",partial:"Partial",features:"features",interface:"Visual Dashboard",onboarding:"Zero-Terminal Onboarding",languages:"15 Languages (IP Detection)",voice:"Native Voice Control",routing:"3-Level Intelligent Routing",zero_knowledge:"Zero-Knowledge Privacy",offline:"Graceful Offline Mode",security_ui:"Security UI (Approval Modal)",context_enrich:"Auto Context Enrichment",quad_fusion:"Quad Fusion Engine",disable_btn:"Toggle On/Off + Voice",compatibility:"Auto Compatibility Detection",dashboard_cards:"Real-Time Dashboard Cards",cost_display:"Live Cost Savings Display",agent_local:"Local Agent",channels_24:"24+ Channels",skills_hub:"13,700+ Skills",multi_agent:"Multi-Agent",open_source:"Open Source (MIT)"},wu={2007504975:"Fill in the new customer's information.","2302290550":"Manage partner applications and accounts","2540171646":"1 && state.step","2907379725":"Extension Layers","2920590826":"You will no longer receive any emails from Thémis.","3237989171":"paradigm shift","3623970196":"Migrate in 2 lines of code â streaming, function calling, vision all work identically.","106e1094c4":"95% Cheaper AI Inference",ad5a2119b2:"Drop-in OpenAI-compatible API. GPT-4o, Claude, Llama, Mistral â all models, one endpoint. From $0.05/M tokens.","8fd3df0d01":"See Pricing","71b1333910":"Try Free","00cdaf2058":"72-Qubit Quantum Power",f9e5c89ef2:"Infernex Quantum â QAOA optimization & QRNG for finance, pharma, and defense. The only inference API with real quantum acceleration.",c81bd1ff38:"Compare vs. Quant",e20d58282c:"4 PILLARS, ONE API",ba9b03469a:"Use the 4 pillars via one API to make your startup",c64248b257:"in your sector","10f517f3c3":`Whether you need cheaper AI inference or quantum-accelerated computing, 191 Infernex delivers both through a single, unified API.`,"2a05a3428b":"Savings Report",a8d89b9803:"Tokens per year",eeaf3cb94a:"Market Cost","5b213d0efe":"Infernex Cost","99b088c045":"Savings","5ac9457fed":"% savings",d08924d375:"saved per year","53efc381a8":"+41 XX XXX XX XX",b3334cb8c1:"â ï¸ Something went wrong","3de10edbe2":"Failed to submit. Please try again.","5cccc7c77c":"Check the proof",f4dde07235:"SOC2 Type II","1149712f9b":"INFERNEX SOVEREIGN TRUST",fc11c14dea:"© 2026 Infernex Corp. â Geneva ð¨ð â",e054569390:"French","10fef62060":"Portuguese","299adc59f3":"Türkçe","2ee663c107":"Themis","5026c69237":"QPU vs Quant","9a47c48c68":"Generative Engine Optimization","3201a9e249":"Separately","63420a0456":"Each product can be chosen individually according to customer needs.","8d091e693a":"AI pack for teams that want multiple services together.","059ff4ce8f":"Complete suite: Quad-Fusion, Quantum 72 qBit and TSU PBit.",ba2f2ff639:"â ANALYSIS IN PROGRESS","31349ef2db":"Infernex.ai: The Quad-Fusion AI API â Quantum, Photonic, Thermodynamics, MiroFish","1d4275b3a6":"Discover Infernex, the fastest, most accurate and energy-efficient AI API. Merging 4 engines (Quantum Q72, Photonic Verde, Tsu PBit, MiroFish), Infernex redefines artificial intelligence. OpenAI compatible. Get your free API key.",a8bd4c815f:"Add a customer",c70d8f1160:"Company Name","3bf846abb5":"Monthly income (â¬)",cecd265776:"AURA â Command Center","268312e14b":"Remote orchestration of every AURA agent installed on every client.","7014482581":"AURA Update â Push to all agents",b64fdfb1cd:"Push AURA optimization to all agents",ff7a08b1ed:"Release notes (auto-translated to client language)","28d8024925":"Improved Stripe webhook handling, faster Workday sync, anti-fraud v2...","08fbb84bb8":"14-day pre-notice sent to every client in their language.","267ef73d03":"Executed in each client's time zone, weekend night only, max 60 min downtime.","8a62471e1b":"Failures auto-retry next weekend with explanation to client.","5877b9b09d":"Recent update campaigns",b6517875de:"All AURA customers (free trial + paid)","91fe10d816":"No AURA clients yet â they'll appear here as they sign up.","69324b94d9":"Automatic Budget Allocation","04f8cf6956":"Active rule:","885ac29a5e":"% of income",a8ee5312df:"Estimated budget:",b4f249fa25:"Income below the minimum threshold ($5,000). No automatic allocation.",ef1f5f25d1:"Allocation in progress...","18e833ce9e":"Run allocation now","84ae7298ff":"Allocation Rules","1a0902d67b":"Allocation History","4a6b997eac":"Recent Decisions","1a085436e9":"No allocation decision recorded","7252edc713":"Search by name or email...",ba9617c188:"All plans","292d37187f":"Previous","0d24441d34":"Delete this customer?","6412cbedee":"This action is irreversible. The client will be permanently deleted.","29b39a43d7":"Quote sent","925e3f110c":"Accepted",a282861979:"Paid","4367ab64a2":"Refused",ea4545f214:"Manage requests and generate access codes",c089d85495:"Generate","0e55f85eea":"No Enterprise requests yet",e48b2af41f:"Create a quote",ec4ea53cb8:"Monthly price (CHF)","8735f6f30d":"Save quote",c9a59447ae:"Infernex G
191EO Engine",a3e3f1fc3f:"Autonomous SEO/GEO Intelligence & Citation Dominance","02cdf5faab":"No data. Run a check.",fb847d384b:"Keyword Gap Radar","91b90a6e68":"No gaps. Run harvest.","45504265df":"Generate Next","14ac9ee931":"Empty. Seed the tail.","6cef1db1bd":"No articles yet.","209c3ea2ed":"No runs yet.","8c2a285cf8":"Complete A/B Testing","506fbd4055":"Growth Channels",fa7f6f4899:"50+ languages supported ⢠Auto IP detection ⢠Adapted hashtags",d0668b6c2d:"Close history","78d7cb3c2a":"No data available",a546e2cd09:"Used","169286fa5a":"Expired","0cfd9157e2":"Loading promo codes...","4f85782198":"Codes generated","44a9196365":"Used","5f49c8556e":"Conversion rate",e46562502e:"Conversion by target plan",ae9634ba04:"No promo code generated","6a9a57d53d":"Promo code history","1c20bc88db":"Discount","2b385dd7df":"Created","418854da93":"No promo code generated at this time",aca173c4a9:"Displaying the last 20 codes on",b014d32248:"Import CSV Prospects","6941dcbe85":"Selected file:","8523c274e0":"added",bf0299a21c:"Evolution of Revenues vs. Ads Budgets","2c43247a76":"Comparison over the last 12 months","2274a53076":"The advertising budget represents on average","6521b591aa":"gross income.",e8799a8d03:"Update...","9f061fe499":"Latest update:","48d47e38e9":"Real time",d3837927c0:"Thémis â Automatic prospecting","2a6201d6e4":"B2B law firm pipeline",a57beb1289:"Generate trim reports.","48680fac2d":"Contacted",ab5002895f:"Answers",b6f782cf16:"No leads â click âEnrichâ to get started.",a1f46d15e2:"White Label Partners","418658ca10":"No partner applications yet","5d5cd268b8":"View Details",e72af1f9fc:"Partner application details","612759ce5c":"Share on X (Twitter)","8d2a893f30":"Share on LinkedIn","81f4074bd6":"Open Source",b36000a2e5:"Network access allowed (no proxy blocking)","774ca7af45":"At least 3 GB of free disk space","0258c91315":"Antivirus does not interfere","69bd0313cb":"Download",f02018cc43:"Security","76dd26a2a0":"Ready",f0d64cdfd5:"Guided installation",a0a1cab636:"5 steps â real end-to-end testing with your local AI",c70e611ead:"Step 1 â Prerequisites & download","19f958f1ba":"Check your system before downloading.",dbc40593ee:"prerequisites not met",cc6140ca7e:"â See Troubleshooting page","72c32a552e":"OS detected:","86956b862e":"Answer Yes/No to each line to activate the download.","0bea5664ba":"Step 2 â Run the installer","0735111487":"Unzip the ZIP and double-click:","23d26f5747":"The script checks Node.js + Ollama, downloads the template, then starts the agent at ws://localhost:18789.",fca6b7dc3e:"Automatic detection in progress...","5f4861a988":"Some browsers block automatic local detection. You can continue without reloading the ZIP.","31ddb7da68":"I launched the agent â continue",cac05fa061:"Step 3 â Model & token selection","67e395dcda":"Local model","0b6ed482b3":"/req (elect)",f37cc56c1e:"Generate a secure token","3b3a89420c":"Run the E2E test",ad265fcb3b:"Step 4 â Actual end-to-end testing",e5965d4df0:"Check: agent â Ollama â model",a3d58b4837:"â AI response validated",af31d6c05f:"First pull pattern may take 2-5 min depending on connection.","84b2b93771":"E2E test validated in","302f6e810d":"Actual AI response:","09b47bb1ba":"Failed test",ed7abe3ce0:"Try again","072d68963c":"Troubleshooting",dc8a5ecda2:"ms with",fa1b121eca:`. 192 All Cloud requests go through`,e2c3602cba:"(Photonics â Quantum TSU PBit â MiroFish 500k agents) for minimal cost.","5e71af9a95":"Open the agent dashboard",e641e114f2:"Download .txt","8994e907db":"See troubleshooting","085269e4c4":"Generating your key...","0b19d22248":"The most gifted AI agent of its generation",ff3d508b11:`Welcome to IAO â The Quad-Fusion AI. Ask anything: code, design, analysis, images, files⦠193 Four engines, one answer.`,"31ce438a38":"Ask anything to The Quad-Fusion AIâ¦",fd4f05c7f7:"No conversation. Click âNew Conversationâ to begin.","0e1da195e5":"Link copied",b422189538:"Share and earn 30% recurring commission.","26aec68de4":"AMY Reward · Partner Program","7cd106b6b5":"30% for life",da308c59bd:"on each sponsored subscription","70a8fb9e40":`
193Share your unique link. Each builder who activates a Matrix (â¬9.90) or Elite (â¬28) plan 194 earns a recurring commission, month after month.`,fb714a3ee5:"Duration",af5e7cb211:"For life","08378ca3eb":"First choose a brand name","3af6ba1fef":"Enter the new name in the AMY field (step 1) before downloading.","496ecaa461":"cloned","172b55576d":"â replaced by","39909e63e1":"Overview","0cc38f8dc6":"Crawl in progress · up to 50 pagesâ¦","314ebb99f8":"AMY · Getting Started Guide","389aa036ee":"Rename the brand (legal)",d9aaa8ce16:"All occurrences of",e8a82a5d25:`have been 195 replaced by a placeholder. Enter your new name â it is applied live to the preview and code.`,"58095ba570":"Connect your domain",a048c83423:"Point your domain to the NEO servers. Automatic SSL in 60 seconds.","7af69bfdee":"Enables one-click payments. Self-configured webhook.",fa5c466291:"Deploy your clone","044d4aba1c":"Download the full React Router ZIP (all cloned pages + your new brand applied).","681466fa7a":"Download ZIP",c309f9a6f1:"AMY rereads the steps","0abbf20a9e":"â Everything is ready",dfc5adf539:"Direct comparison Classic AI platforms vs NEO CODING","420cfa938e":"Functionality","6bb0523fbd":"No hidden fees. No limits. Just your site, in your name.","50517c1233":"AMY HUD â toggle microphone","844e665f1e":"Previous slide","9d636c6a54":"Next slide",bbaeea4b59:"Go to previous page","406b6936d9":"Go to next page",a6bfb486ba:"AI Inference Pricing Comparison 2026: Infernex vs OpenAI",e3ba198701:"ð PRICING COMPARISON â MARCH 2026",ee42d5078f:`Side-by-side pricing for every major AI inference provider. 196 See exactly how much you can save with Infernex.`,"5466780f02":"Calculate Your Savings",b0185bd9e4:"Updated March 2026. Output token prices.",f0592ed6d9:"Why Infernex Is the",a2d98c4fd6:"Start Saving on AI Inference Today","35fe36fa2f":"Free tier. No credit card. 100% OpenAI-compatible.","70009e3de8":"Get Started Free","38695eaa61":"THEMIS",d60f5b0a7a:"Audit Dashboard â Infernex","821a05a493":"Quad-Fusion Audit Result â Infernex",f702d110fd:"Live WebSocket Terminal","6a63d60c55":"AURA â Administrative Universal Response Agent | Infernex",da18ab6c51:"Autonomous engine that detects, analyzes and executes the elimination of bureaucratic and financial waste with zero human intervention.","77ba87203c":"30-day free trial, then 15% success fee on verified savings","3cc0283907":`AURA is an autonomous engine that instantly detects, analyzes, and executes the elimination of your 197 bureaucratic and financial waste to maximize profitability with zero human intervention.`,d8a53b5abf:"What is AURA used for? â in 3 lines",e2a001edc7:"1. It scans",a9aaf9b50c:`your bank statements and your SaaS to find 198 all the money wasted (dead subscriptions, duplicates, unused seats).`,"878d1687f4":"2. He cancels","794ca021af":`automatically in your place â without humans, 199 no call, no form to fill out.`,a96a615829:"3. He proves to you","1d9202dafe":`each saving (cancellation receipt + 200 timestamped snapshot). You pay 15% ONLY on what was collected.`,"9790d3138b":`Difference with Thémis: Thémis = legal proof of documents (Bitcoin/AdES). AURA = execution 201 administration which recovers cash. Both are independent.`,"53b1c28fc8":"See live demo",b1a8548ee8:"â 7-day free trial · 15% success-fee on verified net savings","77958e2487":"Multi-agent parallelism powered by infernex-ultra (decision) + infernex-pro (execution) via api.infernex.ai/v1","04e9128059":"Live Mission Control","81e829d80e":"Powered by Infernex Engine","0df10c50df":"Stop bleeding money on","166b61941a":"No setup. No consultant. No human-in-the-loop. AURA finds it, AURA cancels it, AURA proves it.",d36bc3826a:"0 ⬠today · 7 days free · 100 % automatic provisioning · no code to enter · can be canceled in 1 click",b0198469cb:"AURA Dashboard â Infernex",ac278e3f43:"ð¡ Sandbox mode â Day","4559458ccb":"lost since you connected","36d8e619e8":`. 202 Activate AURA Pro to execute the 94 detected optimizations now.`,"217e0ed8c4":"Stop losing â Activate AURA Pro",f022ec0fcf:"Scheduled AURA updates",aba9e3c43a:"No update scheduled. Your agent is up-to-date.","347d999c62":"SOC 2 Type II","05f8c9c5ec":"GDPR · EU AI Act","73228d25f9":"Benchmark Proof Verification â Infernex","1a6eafb79c":"Themis Verify â",bdaa7b9156:"Check the","35c52aca5a":"The SHA-256 hash displayed on the homepage is deterministically derived from live scores. Paste it back here â your browser recalculates it from the base and confirms the identity bit by bit.","3f0476ad5b":"Proven superiority","9591de2ad5":"The scores below demonstrate â verifiable figures and cryptographic hash to support â that Infernex outperforms GPT-5, Claude 4 Opus and Gemini 2.5 Pro on the MMLU + HumanEval + GSM8K + GPQA aggregate.","03d64d4eb2":"No declarative benchmarks. Each metric can be recalculated in your browser in 200 ms.","9c781e604a":"This page constitutes technical proof admissible for external audit (McKinsey, Deloitte, PwC, KPMG, EY).",e0de48c2c7:"Public Methodology
202· WebCrypto browser · Bitcoin anchor (OpenTimestamps) · RFC 3161 timestamp (FreeTSA) · open source code upon request NDA.",f82004c9e2:"External verifiers notified at each anchor","775cbe7690":"â Papers With Code","5c2bd7ed27":"Submission Metrics","43b5d87f25":"â Hugging Face","3c614adbe3":"Open LLM Leaderboard","45c6698b8f":"â Sitemap Ping",e2b68ba75d:"Each new anchor automatically triggers the notification of these 4 verification networks. Engines and platforms index the URL in less than 10 minutes, increasing the number of independent verifiers.","01f623bd23":"Paste the 64 hex characters","7c304a5e70":"Check","2de9308423":"metric","6c690ddcef":"Hash recalculated","614cb5efa4":"Canonical input (check it yourself:",df2746dc77:"Format: Metrics sorted alphabetically, joined by",c84a478fe1:"Raw data (live from benchmark_results)","04e299a790":"Metric","8c6b7f2067":"Category","9fcfedc2d3":"Bitcoin Anchor Pending (OpenTimestamps Weekly Cycle)","7b5e487eb9":"Download .tsr proof","13c1db6034":"RFC 3161 timestamp pending","13ee887684":"How to check yourself","0d31fd6bdb":"The hash displayed on the home page is calculated in SHA-256 on the live scores sorted alphabetically.","5328826772":"This page recalculates the hash in",e893ded1b9:"your browser",c63846320e:"with WebCrypto (open source) â no trust in our servers required.","1dcfb9faf5":"You can also copy the canonical input and re-hash it in your terminal:",c22e03b3ac:"Bitcoin anchor (OpenTimestamps) and RFC 3161 timestamp (FreeTSA) sign the anteriority â verifiable in CLI:","943fc2b6e9":"ots verify proof.ots","671bccaeef":"Port 18789 or 11434 busy/blocked","4f3a541737":"Antivirus blocks agent or Ollama","56f7362678":"Enterprise Proxy/VPN Blocks Pull Model","5eae9422ab":"Insufficient disk space","7403497f34":"Model not found or corrupt","261bbec7a2":"Return to wizard",f3da6d9e67:"E2E test failed?","079e9f8026":`Here are the 5 most common causes and their step-by-step solutions. 203 Download your installation log from the wizard to share the exact context.`,"5341ec44bb":"Still stuck?","12eec6327d":"Download your log from the wizard then contact support with the attached file.","2ab6915abd":"Install Ollama manually",ebc0856a41:"Cost reduction","3237b39370":"Complete migration","0f0b3824d9":"Vollstandige Migration","4dd7f80e11":"Email Preview â Partner Outreach","990f936084":"ð«ð· French",f096035024:"95% Cost Reduction",a50952aff0:"99.99% Uptime SLA",e09889b9e8:"SOC 2 & GDPR Compliant",acd04c067f:"72-Qubit Quantum Computing",a21444a934:"Enterprise AI Cost Reduction 2026 â Save 95% on Inference","9c2ebacd45":"ð¢ ENTERPRISE AI INFRASTRUCTURE","01049ae384":"Cut Enterprise AI Costs","10bedf66fd":`Same models. Same quality. SOC 2 compliant. 99.99% SLA. 204 Sub-10ms latency. 72-qubit quantum computing. Trusted by leading financial institutions.`,"741e4c7f67":"Request Enterprise Demo",e2357029b2:"Ready to Cut Your Enterprise AI Budget by 95%?","439996c9c5":"Custom pricing. Dedicated support. SOC 2 compliant. Zero downtime migration.","5e45e0a241":"See Comparison vs OpenAI","8f6a7ae826":"Your brand. In every AI answer.","30cb9c7ae2":"Infernex GEO tracks and optimizes your visibility across every major AI platform automatically.","6f3f6851f5":"GEO Dashboard â Infernex AI",d34bff5626:"Hire Quant Alternative | Quantum Computing for Finance 2026","165b7048c5":"ð§ QUANTUM AI vs HUMAN QUANT","7ca0d4769d":"Deploy Quantum AI","3aa0865a85":`72-qubit quantum computing. Sub-50ms decisions. Zero human error. 205 Infernex QPU replaces your next quant hire â at a fraction of the cost.`,"4f2de9e836":"Explore QPU Plans","04c2142a8e":"Quant Analyst vs","1d43274282":"What Your QPU","24e38cb931":"Pricing Models","09eb875049":"Quantum Monte Carlo",b155959fb2:"Your Next Quant Hire Costs $1,999/month",d94f44dd87:"72 qubits. Zero bias. Infinite scale. Start today.","3f064a2c10":"See QPU Plans",ee2936be8e:"Powered by Infernex Quantum","793e4619ea":"Translation Audit","3845c75735":"QAOA quantum intelligence to translate the entire site into 15 languages with 99.2% accuracy","18251c31a4":"Quantum Audit I18N",d66d6c5cab:"critical pages and generates all translations","5f58abea51":"Quantum analysis in progress...","05daecb702":"Audit completed","6d94eb9da7":"Channels found","8b055284c0":"Accuracy","1b1f4c8400":"Download JSON translations","9e7e52fd04":"Quantum optimization algorithm to ensure semantic consistency across 15 languages",edd119ffc7:"Model augmented by Infernex infrastru
205cture for accurate contextual translations",ea2aba89b9:"Structured JSON ready to integrate into i18n files without manual editing",bfecedb9cb:"Production note","31fcf47a29":`This demo interface uses placeholder code. In production, integrate the actual contents of the files 206 or use the edge function directly in your CI/CD pipeline.`,"1b699326da":"IAO: The Quad-Fusion AI by Infernex | 4 Engines in 1","90e1a5e6b2":"CAE Chat Widget","2fd64cf3f2":"Sign in Required",bead507926:"You need to be signed in to access IAO.","0c02272057":"IAO Plan Required","5d652ef151":"Upgrade to the IAO plan to access The Quad-Fusion AI â the most gifted AI agent of its generation.",ca47c555a2:"View IAO Plan â $17/mo","5e33bc9f4b":"IAO Chat â The Quad-Fusion AI",ae17abb98a:"Try IAO â Free Demo | Infernex","29e0509ebe":"Free trials ·",fc91a77909:"Welcome to IAO Demo â Try The Quad-Fusion AI for free.","1f3afca2d0":"free messages · No account required · Code, design, analysis, imagesâ¦","562c52a263":"Unlock unlimited IAO for $17/mo",f9f2ac6b0c:"Create Free Account",aeee86f93e:"Try IAO â ask anythingâ¦","408d6ce965":"Upgrade for unlimited",be3d6dac81:"IAO Chat Widget â Embed AI on Your Website",a4d10b8a3a:"Infernex.ai: The Quad-Fusion AI API â Ultimate OpenAI Alternative","4271ff9060":"Welcome to the Age of AI API","559212f744":"Infernex merges 4 separate engines into one","345744a870":"OpenAI-compatible AI API","9272b7fb6c":", treated in this order:",a4075a13c3:"â ultra-efficient photonic inference","93c0c65efc":"â sub-millisecond latency",c47252c8f9:"â advanced thermodynamic precision","75e2bb7f97":"â distributed agent consensus",d26c7ec57e:"The result is not a simple",b0c3e3ac1e:"AI performance improvement","77e4dbd8ea":"for the","2b3155d91c":"AI application development","46fa8f8ffc":"Scroll to benchmark â test the fastest and most precise AI API","7b38207271":"Discover the Infernex Advantage: Test the Fastest and Accurate AI API.",c3b09b6209:"benchmarks run today","78d1143876":"Complex Prompt â AI Bioinformatics Simulation","58a8b27d21":"Enter your own prompt or use the default one below:",e30229f2e7:"Legacy AI API","675071038b":"â NEW · LEGAL AI",d1ee7c9f8f:"â Your sensitive documents, analyzed and certified in 1 click","46bfb56d81":"Law firms · HR · Accountants · Notaries · Compliance · Doctors",dfae5a2b6e:"Drop off. Themis","7fadfe81c5":"number in your browser","1cf366dbee":", anchor on",f998eeac62:", timestamp (RFC 3161), generates IA summary and GDPR certificate.","3abb51cf92":"100% publicly verifiable · â¬0 hidden extra costs.","2e4679a685":"Market standard",e004ba0ce4:"Themis Infernex",abd473501f:"â Acceptable in 80+ countries · 15 native jurisdictions",c7cacb08a2:"eIDAS AdES (EU 27) · ESIGN Act (USA) · UK eIDAS · PIPEDA (CA) · SCSE (CH) · ETA (AU/SG/AE/JP) â self-authenticating Bitcoin proof (FRE 901/902).",a2226e72b1:"â Used by law firms, notaries, private banks and HR services in Switzerland ð¨ð",a6cd70f4ad:"â¶ Get started for free","303745ffa1":"Check a hash â","1bf6ab4984":"Free for life · Without credit card to start · Unsubscribe in 1 click",b0edf7423a:"Hidden additional cost",e2d1440c04:"â Proven compliance","5fb38a44cf":"6 months Pro",f50ef7e73e:"for the 20 founding firms.","76fefb9396":"â Exclusive Thémis offer ·",e1715a4b03:"with the site's monthly promotions.",a9d00ee9c5:"The AI Power Tech Leaders Need.",fd1f5ef575:"Infernex is not a general AI API. It is architected for organizations that demand minimal latency, flawless accuracy, and maximum efficiency.","5045c1d063":"Advanced AI Architects",aeb51cbe93:"Universities & Research","2167a5bb6d":"Defense & Regalian Sector","3da016f2eb":"Agencies & Creative Studios",a27c45cafd:"â¡ NEW · AI CODE STUDIO",a411c44091:"âCanvas Domination","2f9a8babba":"The most advanced code creation environment. Infinite Canvas, AMY avatar, Exodus module (AI cloning) and integrated Quad-Fusion routing.",f8e36bc40a:"Infinite Canvas Deep Dark Matrix","7ceade1ca8":"Avatar AMY â Animated AI Assistant","6ebeca2053":"Exodus: Clone & rebuild any site","85cd964539":"â¶ Launch NEO Coding","3d44e9a625":"Free access to Canvas · Matrix plan for export","2f5aa34cb4":"Welcome to NEO Coding. Drag the nodes onto the canvas to compose your architectureâ¦",fe9e295c98:"Unleash the raw power of AI. Integrate the Infernex API today.","7b3be90126":"GET YOUR INFERNEX API KEY","60437113a5":"Free Tier Included â OpenAI SDK Compatible","8530f3979b":`Direct replacement solution for OpenAI API. Start a new AI project or optimize your existing infrastru
206cture. 207 Ideal for`,"4bd98d658e":"AI developers",bd731e1dcc:", the","002a9e3352":"and the",ac7109aa5f:"investors looking for the next AI revolution",b85979c770:"Get Your API Key",f6271a8cb8:"Enter your email to create a free account instantly.","4a2f5c8a04":"Please fill in all required fields",e58ef10032:"Please use a professional email address (no Gmail, Yahoo, etc.)",e9f36d65b0:"Something went wrong. Please try again.",fa9d816dce:"Thank you for your interest. If your profile matches, we'll send you the private pitch deck link within 24 hours.","6957b0b690":"Back to Home","433e5133e7":"Our pitch deck is shared privately. Fill in the form below to request access.","5d7ae886c8":"Your venture capital firm or organization",cc716fe4e3:"Tell us about your interest in Infernex...","0fa1209548":"â Back to Home","0b99a60857":"Source: GitHub Trending, 2026","22f1f6a173":"MiroFish Integration Guide â Infernex API Setup",a217057cde:"MiroFish Integration Guide â Connect to Infernex in 3 Steps",e0884ab667:"This guide explains how to connect MiroFish to the Infernex inference API. MiroFish uses the OpenAI SDK format. Infernex is fully OpenAI-compatible. No code changes are required â only two environment variables need updating.","68887e44ba":"MiroFish installed (git clone https://github.com/666ghj/MiroFish.git)","55cc31283b":"An Infernex API key (free at infernex.ai/signup â includes 10M free tokens)",abe6e5f4e9:"A Zep API key for agent memory (optional but recommended)","9a68566a88":"Step 1: Clone and configure MiroFish",dca17f9ce4:"Step 2: Edit your .env file","74b4573018":"Step 3: Install and run",a00f22f7ec:"MiroFish is now running on Infernex inference.","9b819ee1cc":"Recommended for full MiroFish simulations. Best balance of quality and cost at $0.04 per million tokens. Supports 2,847+ competitor agents.","2cfff1b265":"Recommended for high-volume or real-time agent tasks. $0.02 per million tokens. Best for trading bots and rapid simulations.",c85b2f2dde:"For simulations requiring image or document input. $0.06 per million tokens. Full multimodal support.","8edb99f178":'Error: "Invalid API key"',"54ae8c7ea5":"â Verify LLM_API_KEY is set correctly","5833aee87c":'Error: "Model not found"',"971046a174":"â Use exactly: infernex-ultra or infernex-fast",a7164c8d3b:"â Verify LLM_BASE_URL=https://api.infernex.ai/v1",e2340b97cc:"â Switch to infernex-fast for smaller agent tasks","82b73d28cb":"High token usage","920e49a929":"â Enable prompt caching in Infernex dashboard",cb07ea52bd:"Integration landing page","7438c1fa25":"Estimate your costs",a771bb0e72:"Cost comparison item",e4fb2fe140:"Quantum boost addon",bf14a89a9e:"Get Free API Key",b857e1258d:"MiroFish + Infernex FAQ â Everything You Need to Know","85e46d2385":"MiroFish + Infernex FAQ","60bfa74899":"Everything you need to know about running MiroFish on Infernex.",c8eef1aaec:"Enters a URL to clone.","7ca3715485":"Demo Mode","5d66c31c4a":"Exodus completed","1db1849934":"Exodus Failure","6083245db2":"Matrix plan required","19da8f4ebb":"Enables the Matrix plan to export the code.","34266c25e8":"Connects an external domain with the Matrix plan.",ffbe8c7d17:"DNS ready","0ff8228206":"NEO CODING â Canvas AI Quad-Fusion on Infernex",e774bc9a67:"Geo ·","21547db102":"Toggle AMY voice","1537056ef2":"Demo","4cfdae3afb":"Exodus · Clone URL",c786c51ab6:"Clone any site quickly with its URL, or create your site from scratch by chatting with AMY.",cbd41dd2c9:"Create site","7a4fbb2f12":"About Exodus","1df7f90b1b":"Auto SSL Matrix plan required.","23d06efa57":"Legend","5599e58815":"Export, deployment, domains, priority support.","86d31c4351":"Download started","77275b9d62":"NEO creation · new project","315aadff53":"Generated React/Tailwind code","141f76f396":"Plans under the Canvas","32b7bcfd8c":"â URL cloning â","731ca68093":"10 pages max","4ff9a7247a":"â Free React/Tailwind export","8560e9926c":"â Deployment & custom domain",b4c43cc484:"⢠Phrase âPowered by AMYâ required","039935559f":"No obligation · Cancel in one click.","995024f52f":"â Everything from the Matrix plan",cdfdfd5be7:"Complete site",de250f702a:"â up to 50 cloned pages",f86043ed9a:"â Complete white label (signature removed)","48693b4409":"â Personal subdomain & auto SSL","4d9f08c42a":"â AMY priority support",ac3aa7738b:"Powered by AMY AI | Neo Coding",e8a28bc7b4:"Install the SDK (unchanged)",f883eb90ea:"Change the Base URL","128a81a6d1":"Use the Same Code","63c458d215":"GPT API Cheaper Alternative | OpenAI Migration Guide 2026","41a66a3038":"ð MIGRATION GUIDE",f07bb014f4:"Switch from OpenAI in","18883056e4":"Change one line of code. Save 95% on your AI inference bill. Full compatibility guaranteed.","285df7d364":"95% cost reduction",f5cde7592e:"1 line of code","5564a21f0a":"Get API Key Free","825e31404a":"3 Steps to",be478da25c:"Stop Overpaying for GPT Tokens","39b61d7df8":"Join 500+ companies that switched from OpenAI to Infernex. Average savings: $8,000/month.","9fa8a0a5d8":"Migrate Now â It's Free",f803cde3e1:"Access OpenAI's most powerful models at a fraction of the cost through our optimized rout
207ing and caching infrastructure.","5701ee97f9":"99.99% SLA Guarantee","33b1b71aa3":"Enterprise-grade reliability backed by our multi-region failover architecture and proactive monitoring.","10c34ef1ec":"Your data stays secure with end-to-end encryption, zero data retention, and full compliance with enterprise security standards.","5f308ecf42":"Global Edge Network","342a4c7d58":"Sub-10ms latency worldwide with inference endpoints deployed across 12+ regions for ultra-low response times.","6b00667133":"Real-time dashboards with token usage, cost tracking, and model performance insights for every API call.","48c8488900":"Dedicated partner success manager, private Slack channel, and 1-hour response time for critical issues.","4b5a7e7f29":"Volume Pricing","26a908a8b4":"Partners â Infernex | OpenAI Infrastructure Partner",d008a5fb92:"Partnership in progress","1be7542d99":", optimized by","94e5fc6a3a":`We're building the fastest, most cost-effective gateway to OpenAI's models. 208 Every token through Infernex is an OpenAI API call â at 95% lower cost.`,f0f85d23e5:"Try the API","84b5cc39be":"Become a Partner","4023ee5944":"Our journey to becoming an official OpenAI infrastructure partner.",d1dd2c082a:"Why Partner with Infernex?","8c96fadb55":"Get all the power of OpenAI's models with infrastructure built for scale.",e13727f017:"Switch in 2 Lines of Code",abb701497e:"No lock-ins. No breaking changes. Just change your API key and base URL.",ea51f3ee21:"View Pricing","32e56a0e13":"AI Inference + GEO Optimization + Quantum Computing â One Platform, Three Revenue Engines.",eb285a7cdc:"3 Products Live","6664b4a873":"AI Costs Are",b2c713a7c1:"Companies spend 60-80% of their AI budget on inference. Meanwhile, zero visibility in AI search engines destroys organic traffic.","1514d3acef":"of requests don't need GPT-4",f8c56ebf73:"visibility in ChatGPT, Gemini, Perplexity","82f121e54e":"avg runway lost to inference bills",ea1306bc89:'"Our OpenAI bill was $52K/month. At that burn rate, we had 6 months left." âAI Startup Founder',a53726e3c1:"Infernex analyzes every prompt in real-time and routes to the optimal model â same quality, 95% less cost.","46359c71dd":"â Zero code changes needed · Try it in our live Playground",aabda4eca3:"How It Works",bdf600738d:"â Premium models",e3a0bfb4e8:"OpenAI SDK Compatible","84b7f6239c":"Streaming & Function Calling",a2a1c65b67:"Each product targets a different $B+ market with independent revenue streams.","05dae75f98":"AI-Powered Test Data Generation","8a5f8b9f2b":"GDPR-compliant synthetic data","4678e62c78":"SQL, JSON, CSV, XML output","1a0ae4ec7e":"Data market testing",d15173a598:"Get cited by ChatGPT, Gemini, Perplexity","26dadf7168":"AI article generation + CMS publishing",f60948b941:"Citation tracking & gap analysis","2c7f558cbb":"SEO/GEO market by 2028","1f4247e76a":"QPU-Accelerated AI Inference","17b0ba1764":"QRNG for true randomness","8984217690":"QAOA portfolio optimization","968e39b445":"Quantum computing market by 2030","7559575e4e":"Models, One API",ec93773bd7:"Infernex sits above providers â routing each request to the optimal model across the entire ecosystem.",f35e326885:"Infernex Routing Layer",d9aac2b478:"Languages, Day One","2e80c1c19e":"Full platform & marketing localization to capture global developer markets from launch.",c7d18d5616:"Arabic & Hebrew support","6c15f93db8":"Localized comparison pages",ca5905202c:"Three overlapping markets, each growing 30-50% YoY. Every company using AI is a potential customer.",f1eab699ae:"Core inference routing",fc07705977:"Quantum computing by 2030","137e710828":"Source: Grand View Research, McKinsey AI Report 2025, Gartner","94be466ff7":"â¡ Inference Platform","43e3ea66ac":"â Full OpenAI API compatibility",b3cb9262fd:"â Smart routing engine live",f55b17802f:"â Live Playground","558442052f":"â White-label partner program",f31799f9c1:"â Stripe billing integration",e825e85b23:"ðGEO Platform",faaf34ce21:"â AI quote scanning",c40f07038f:"â Self-generated article",f36d72963a:"â Self-publishing CMS",eb2e709b61:"â Keyword gap analysis","8d4300fb55":"â Trial system live","8431e211ef":"â 15 live languages",cb42b25b28:"â 10+ comparison pages (vs OpenAI, AWSâ¦)",a69c544a18:"â Automated blog & content pipeline","0198f3a338":"â Partner outreach automation","2d7c5fb7b1":"Usage-based + SaaS subscriptions across three products. High margins from provider arbitrage.","5b9b86f1e8":"$49-499/mo · Test data generation","910e553b07":"$99-999/mo · AI citation optimization","152d702f5f":"Blended Unit Economics",a80e94b9c0:"Gross margin 70-95% across all products",b0499929dc:"Our moat: Multi-product platform
208+ intelligent routing + semantic caching + quantum QPU + 15-language network effects","03964469c5":"A global leadership team across 3 continents.","4ae2c5fafd":"To scale three products, accelerate GTM across 15 markets, and capture the AI infrastructure opportunity.","3af1b9b685":"Use of Funds","75eb6278dd":"Let's Build the Future of","654de5373a":"8 AI Tools, One API â 95% Cheaper | Infernex 2026","9fe45ad51e":"Google Willow Quantum","3219157ae0":"IBM Quantum Platform (Qiskit)","74903780fd":"Infernex QPU Enterprise â $27,440/month â¦","452e3ec945":"curl -X POST https://api.infernex.com/v1/quantum/run \\","90035e830d":'"Authorization: Bearer inf_sk_..."',d29f6327e9:"72 Superconducting Qubits",a93372ba56:"QRNG Random Generation","329d3b3b64":"Sub-50ms Decision Latency","30cc986f72":"Quantum-Accelerated AI Inference 2026 â 72-Qubit QAOA & QRNG","905cbde37d":"AI Inference Powered by","976b661935":`The world's first AI inference API with integrated quantum computing. 209 QAOA optimization. QRNG random generation. Infernex Quantum â 72 superconducting qubits.`,"5662bbb256":"QPU vs. Human Quant",dcd808a0b2:"Quantum vs. Classical",a4fabda13c:"3 Lines of","9e896b587e":"# Python â Quantum Portfolio Optimization","59dd748218":"The Quantum Advantage Starts at $1,999/month","6727ba0164":"72 qubits. QAOA + QRNG. Classic fallback included. No PhD required.","68390fc2f0":"View QPU Plans",ac49883007:"Quantum Ads Dashboard","5ded00b681":"No campaigns found. Add campaigns in the admin panel.","26c8a16035":"Quantum API Documentation â Infernex",d0333365c2:"Infernex Quantum API v1","13ea0ac8b0":"Complete reference for QAOA optimization, QRNG random generation, and quantum-accelerated AI inference.","0113ad8af6":"Get started with Infernex Quantum in 3 steps:","00fbd4edac":"1. Install the SDK (Python)","7d5478a820":"2. Initialize the client",f9cfe5de9e:"3. Run your first quantum operation",c2150cc82c:"All API requests require a Bearer token. Get your API key from the","0f7ade123e":"Quantum Dashboard","5577b7111a":"QRNG â Quantum Random Number Generation","0a922bbab0":"Generate true random numbers from our 72-qubit quantum processor. Uses quantum superposition for true randomness, with automatic classical fallback.","33bffe717a":"QAOA â Quantum Approximate Optimization","7e2f608569":"Solve combinatorial optimization problems using the Quantum Approximate Optimization Algorithm on 72 qubits.","24dc21ed18":"Supported Problem Types","440a47ce83":"â Asset allocation across securities",cdd923d3ab:"â Graph partitioning","66e22b4a58":"â Route optimization","42342ab2b5":"â VaR/CVaR stress testing","7ad6b21e8a":"Quantum-Accelerated AI Models",dc019d5693:"Standard AI inference with quantum optimization for routing and caching.","240f71bbf0":"Get Quantum Access",b7dcf0b694:"Drop-in API Replacement","55a85d030c":"Instant 95% Cost Reduction","368752690a":"How to Reduce AI Costs by 95% | Enterprise LLM Optimization",c5825718a2:"ð° REDUCE AI COSTS BY 95%",edf9e4e44e:"Stop Overpaying for","4ec63aeb0a":`Same models. Same quality. 95% cheaper. Infernex is the drop-in OpenAI-compatible API 210 that cuts your AI spending from day one.`,"3f321fc3a8":"Start Free â No Credit Card",c16a994fa5:"How Much You'll","861c0c6ef8":"You Save",fbc28ea207:"Ready to Cut Your AI Costs by 95%?",fa35c57c39:"Free tier included. No credit card required. Migrates in under 2 minutes.",b684ee968b:"Start Free Now","91dd67e4ec":"<10ms Cold Start",bc1ca6ffcc:"Pay Per Token",ea67557618:"99.9% Uptime SLA",c5d0dc479a:"Serverless AI Inference | Scale LLMs Without Infrastructure","7fb2ba5c86":"âï¸ SERVERLESS INFERENCE","66f32ca7e8":`Zero servers. Zero GPUs. Zero cold starts. Deploy AI inference that scales from 0 to millions 211 of requests â pay only for what you use.`,"3d251afaec":"Start Free",bd14dbd797:"Why Serverless with","663c90148f":"# pip install openai","421c77cb22":"Ready to Go Serverless?","691ac537e0":"Start with 1M free tokens. No credit card required.","1f1bfa030a":"Infernex AI â Quantum Supremacy War Room | Live Proof",db5c1806e0:"Quantum Translation Engine","13cfa6b9b3":"Complete 9 incomplete languages (~11,000 lines) with Quantum-powered AI (Gemini 3 Flash)",ea396713d6:"Batch Translation Generator","6e1410aeaf":"Generate complete translations for all 9 incomplete languages",cba0207798:"Generate All Translations","10cdaa4b19":"All translations generated successfully","9ffdb95250":"How
211it works","5f097ccc6d":"Quantum AI Translation:","32eb452d03":"Uses Gemini 3 Flash via Lovable AI Gateway","395c4a500b":"Generates all 9 languages sequentially (2s delay between requests)","6c3535e1d4":".json and replace in","5a4a9f1330":"QAOA-optimized translation maintains context and technical precision",d86ba0629a:`Each language takes ~30-60 seconds to generate. Total process: ~8-15 minutes for all 9 languages. 212 Files are downloaded directly to your browser - you'll need to manually copy them to your project.`,f519f7a5f5:"Infernex Ultimate Supremacy | TSU PBit All-in-One | 99.7% Accuracy","23cbc321c5":"⦠Enterprise Only",dddd79e58d:"Unsubscribe â Infernex AI","926548112c":"Confirm Unsubscribe",a99221a69d:"Back to Infernex","8d5455c6ea":"You've been unsubscribed","9d8925ac6a":"You won't receive any more marketing emails from us. You can still receive important account notifications.","0b0d82bd94":"Widget API key missing. Add","2b0d274cef":"to the URL.","695d4f50e2":"Powered by Infernex",ac5ca4d121:"Cheap AI Inference: QPU Optimization & Cost Reduction","3a8a681abb":"How to Use the OpenAI Python Client with Infernex","49feff61fb":"Migrate in 2 lines â same SDK, 95% cheaper.",f1202647f4:"AI Inference as a Service: Complete 2026 Guide","1865e901d3":"Managed inference platforms, costs & migration.","6380d6aeb6":"Are OpenAI & Anthropic Losing Money on Inference?","4a06bf03ef":"Real GPU cost breakdown & what it means for you.",a9974a6cef:"LLM Cost Comparison 2026","2b133300d0":"GPT-5, Claude 4, Gemini 2.5 â full pricing table.",e4492b4fb3:"What is AI Inference?",be955e4805:"The complete guide to AI inference.","64d21ce130":"Enterprise LLM: Reduce AI Inference Costs for B2B",cb7b943421:"LLM for Enterprises",ab76a20cd0:"ð¢ NVIDIA TensorRT",e14876b2fe:"ðµ Intel OpenVINO","159193d67c":"ð¡ONNX Runtime","826d6d7d15":"Google Cloud Blog","20e337f1ff":"Up to 2000% (95% cost cut)",b2e3f26308:"8 min read","2336db499a":"Best OpenAI Alternative 2026 â Save 95% on AI Inference","600920a065":"Infernex vs AWS Bedrock 2026 â Save 90% on AI Inference","678214b9bb":"ð Related Articles","8d1b838c91":"Why managed platforms beat self-hosted AWS infrastructure.","6a747b29a2":"Real GPU cost breakdown and what it means for your budget.",aa68945e4f:"Full pricing table across all major providers.","1233b1b48f":"Managed inference platforms, costs & migration for enterprise.","4b75c45eef":"Google Vertex AI","334aceff51":"Migrate in 2 lines of code â same SDK, 95% cheaper.","9680330495":"LLM Cost Comparison 2026: GPT-5, Claude 4, Gemini 2.5",f4234ebb7d:"Infernex uses Hugging Face as a provider. Compare how Infernex's routing layer optimizes HF Inference API costs by up to 90%.",f625ebcc41:"Infernex vs Hugging Face â Complementary, Not Competing",bfdd32af42:"Not competitors â Infernex uses HF Inference API as one of its providers, adding intelligent routing, failover, and 90% cost optimization on top.",aab0855d4d:"How They Work Together","280785e1fc":"Sends request via OpenAI SDK to Infernex API","2044f9d8a0":"Classifies complexity â picks optimal model & provider","433e832760":"HF Inference API","119588cfc8":"One of 10+ providers â Llama, Mistral, StarCoder via HF",d382b7cf18:"Models accessible via Infernex (incl. HF)",eedaa4b391:"Cost savings vs direct HF Inference API","3e02987f98":"Total latency with smart routing",b8e1a49f21:"Why not use HF Inference API directly?",c3f9363894:"You can â but you'd miss these advantages","0802d63182":"No intelligent routing","655776ffa2":"No cost optimization","9e67f11daf":"No unified billing",d2d39f68d6:"Access 100+ Models Through One API",ebaa2a05a8:"Including Hugging Face, OpenAI, Groq, and more â with intelligent routing and 90% cost savings.","0980122bd6":"Start Free â 50K Tokens/month",cbf9c23d52:"General-purpose reasoning and text generation",e0327114a2:"Real-time speech recognition, 10x faster than competitors","14456a113d":"Multimodal document embeddings for RAG",c656eb72cc:"Cross-attention reranking for search accuracy","1ede0c33a1":"Infernex vs NVIDIA â Run Open-Source Models 70% Cheaper",f94f3d35d7:"CES 202
2126 Open-Source Models","4d42e73980":"Run Nemotron, Clara, and Cosmos at 70% lower cost with 10x better latency",d579f4dcb3:"NVIDIA Open-Source Models on Infernex","650cbaa9f5":"Use the same OpenAI-compatible API for NVIDIA models",dfd8588106:"Why run NVIDIA models on Infernex?","153d5e82e5":"Get the power of open-source AI without the infrastructure headaches",ae5cf35a8e:"70% Lower Costs","172d032233":"We optimize inference with quantization and smart batching to slash costs","82c7fe5d31":"No GPU Management","460807888e":"Skip the H100 provisioning - we handle all infrastructure for you",e08f61b8f1:"Edge deployment and optimized kernels for <10ms response times","5265fd5df8":"99.99% SLA with automatic switching to backup providers in <100ms","28c544da29":"Real GPU cost breakdown â including NVIDIA's latest B200 numbers.","164517b735":"Why managed platforms beat self-hosted NVIDIA infrastructure.","74fcd04dbe":"Ready to run NVIDIA models at 70% lower cost?","292dfc80e6":"Start with 50K free tokens/month. No GPU setup required.",debd74544c:"Start Free Trial","2fe04d5468":"Full pricing table across all major providers with real-world use case calculations.","7038d90374":"Partner not found","687298349a":"à THEMIS â Legal AI eDiscovery","70e4723045":"Documents processed","01e7e22d3c":"Hours saved",cfe27d5d05:"COMPLIANCE & SECURITY","3cdedc825b":"Get started with Thémis",cb02d1695e:"Same preferential rates as",eaa129af92:"BOOK A DEMO","675434128b":"Bates Immutable Stamping",caf7c0148f:"Automatic Conflict Check","602dd1340b":"15 native jurisdictions",eff9b5523d:"Certificate of destruction",a02ab3e6a6:"OCR + AI summary","9073dbd67a":"Thémis â Legal AI eDiscovery Trust-Grade | Infernex",acd90981a7:`Bates immutable stamping, multi-jurisdictional conflict check, OCR + AI summary, GDPR certificate of destruction. 213 Designed for demanding firms â from sole practitioner to Big Law.`,ef25a486f8:"â¶ DISCOVER THEMIS",f10df0acbf:"100% AUDITED â TRUST CENTER",d568c55f4d:"Without credit card · GDPR · ISO 27001 ready · Free public verification","80c25843a7":"Scroll down for more information",d2ad0a3241:"Scroll · More info","50cc2687ed":"Compliance",e8d5505cc9:"â Regulation (EU) No 910/2014 â with native Bitcoin anchor via OpenTimestamps. No hidden costs: open standards, free public verification.","3cfd36b40a":"From upload to proof","12b43b11e7":"LEGAL RECOGNITION","4c1b6b842c":"Admissible in","68c5c6733b":"Thémis combines eIDAS AdES, Bitcoin anchor (OpenTimestamps) and RFC 3161 timestamp (FreeTSA) â an international Trust-Grade standard immediately enforceable in the courts of the main world jurisdictions.",f4f5f9eecf:"Thémis vs DocuSign · Yousign · Skribble","08675a9617":"7 decisive technical criteria. No competitor offers the combination of Bitcoin Anchoring + Zero-Knowledge + free public verification.",b6e65dc7d0:"Criterion",d054adccff:"Legal note:","0c4fb8e602":"Thémis issues signatures by default","1085c7f7ad":"AdES (Advanced)","29e8d0fd04":", admissible in 80+ countries. For acts requiring maximum probative force (automatic legal presumption), the plan","462f2f706c":"30 QES Swisscom signatures",aab8c761e0:"â equivalent handwritten signature according to SCSE Art. 14.",be4fa2be7a:"4 shots, zero surprises","1c2ba201c4":"Monthly rates · VAT extra · No obligation",c8bbd1fe72:"Recommended","45081fe1e9":"Public Bitcoin Verification",c404cc01eb:"Each document is anchored on the Bitcoin blockchain via OpenTimestamps + RFC 3161 timestamped by FreeTSA. Anyone can verify the authenticity of a hash, for free, without registration.","1701f6f9a1":"Verify a hash",b635b904e6:"Take back control of your eDiscovery.",ad41f42262:"START THE THEMIS TRIAL","52b9986d4e":"Anchor Partner â50% · Without credit card · Unsubscribe in 1 click","7c31f5ffac":"â Themis plans are not","600bd2e64f":"with the site's automatic monthly promotions.",ad2e070db7:"Thémis Dashboard â Your sensitive, analyzed and certified documents","721f9a23d8":"Thémis · Customer dashboard",eb956a73a0:"Your documents,","79a720ffca":"analyzed and certified","2bad55897b":"Submit your PDFs, contracts, HR or accounting files. Thémis performs Bates cryptographic stamping, conflict check and AI summary. Each operation is recorded in your private audit trail. Only you have access to your files.",ff8b9c7905:"â Your files are encrypted","5d93504ff5":"in your browser (AES-256-GCM)","63cc6c45ef":"before uploading. SHA-256 hash anchored on","355528a37b":"Bitcoin via OpenTimestamps","630f2f32c5":"(free, public). The key is downloaded automatically â","1b8acabcb8":"keep it, without it the document is irrecoverable","7e8488745f":"See the plans â","944289836c":"Encrypt · Upload · Anchor to Bitc
213oin","470d0bdee8":"PDF · Word · Excel · Images · TXT â 50 MB max â encrypted in your browser, never in the clear on our servers","4441920876":"A unique AES-256 key per document is generated and uploaded. Keep it â Zero-Knowledge = we can't get it back.","678f21b2cd":"Your documents (","883fb76e40":"â³ Refresh","8baf2ab47b":"No documents yet. Submit your first file above.","446f18fb32":"AI Summary",cf6ba37b8b:"Themis result ·","9314865b61":"â Close","410755bf54":"ð Zero-Knowledge AES-256 · â¿ Bitcoin Anchored · â± RFC 3161 · â eIDAS Advanced (AdES)","27e98fcd72":"Verify a document publicly","8a9702a409":"Thémis â Public verification of documents | Bitcoin Anchored Proof","62f3f30dad":"Public Audit","788bd60a71":"Bitcoin Anchored · eIDAS Advanced",a58fa1feb4:"Check a","632fa232ab":"Paste the SHA-256 hash of a document to confirm its Bitcoin anchor (OpenTimestamps), its RFC 3161 qualified timestamp (FreeTSA) and its registration in the Themis registry.","8afe5df428":"SHA-256 (64 hex characters)",eeb9edeed3:"e.g. a3b4c5...64 hex tanks","056288bdb8":"Authenticated document â","691858179d":"Anchored it","19621d6e2e":"Check at opentimestamps.org","4da0bf236f":"Unanchored","425e1b82d3":"Not timestamped",e00a34fffa:"â Client-side encrypted document. The Thémis server has never had access to the unencrypted content.","2ac428deee":"How it works",de8f0d5f51:"When uploading, the hash",baaf1321ff:"of the document is calculated in your browser.",d59d023b27:"This hash is subject to","7423325f5c":", which includes it in a Merkle tree anchored on the","2c61e09827":"At the same time, a",c35ee84f5d:"qualified is issued by","825b32b600":"(Independent Timestamping Authority).","866ed33cb7":"Anyone with the hash can verify the authenticity here, without accessing the document.",b151e7049c:"Compliance:",f6043c5ce2:"eIDAS Advanced Electronic Signature (AdES)",bf62ba2e56:"· Regulation (EU) No. 910/2014 · GDPR Article 17 (right to erasure with cryptographic proof of destruction).","26ff02d725":"Thémis Trust Center â Anonymized public audit","15ad02d540":"â Back Home","65677523a2":"Public crypto audit",fc3a6c3005:"Each Thémis operation is traced with a verifiable SHA-256 hash. No customer data appears here â only the anonymized evidence.",c032081041:"LATEST OPERATIONS","9c1ee1dbc4":"No events recorded yet.",f620f0ed15:"Processing in progress...","825c3b10a4":"UNSUBSCRIPTION CONFIRMED",f3b73b65b3:"The link is expired or incorrect.",d559772388:`t("amy.status.ready", "AMY · prête")); 214 const [firstName, setFirstName] = useState`,"828eca20f9":`(null); 215 const awaitingNameRef = useRef(false); 216 const pendingPlanRef = useRef`,"5a1dd81acc":`(null); 217 const speakingRef = useRef(false); 218 const conversationRef = useRef`,bd71d196ea:`= 2 minutes (computer sleep / long tab switch) 219 if (awayMs`,e95d498e97:"Infernex","493727d622":"Groq","1551e58e6e":"Together AI","1bf43af666":"Anyscale","04558bc4c6":"NVIDIA Nemotron","2eb57b16b8":"Infernex AI",c4f8b577df:"San Francisco · Dubai · Geneva","4027cd0f87":"$0.50/M",a8bcd6a68a:"number 1 worldwide",d05999cdf5:"Calculer","54e601c774":"Tokens/an",fcecdc9f3d:"Exporter JSON","794dcffeb2":"â ï¸ Something went wrong",b873c33c1c:"Refresh page","65572beb3c":"500K tokens",da73f7757c:"[email protected]","24662d6a02":"[email protected]",db7206b605:"www.infernex.ai",b981fcf8a3:`: 220 'isRoute' in link && link.isRoute ?`,"56a2556fc6":"'isRoute' in link && link.isRoute ?","89c9b65356":"X (Twitter)",dd84425b72:"LinkedIn",f911e414cf:"GitHub",b1eba1736c:`#MachineLearning #LocalLLaMA #deeplearning #ArtificialIntelligence #LLM #ChatGPT #OpenAI 221 #Singularity #StableDiffusion #LLMDevs #ClaudeAI #Anthropic #LanguageModels #GPT4 #RAG 222 #artificial #learnmachinelearning #MLOps #NLP #GenerativeAI #AIInference #DeepLearning 223 #ProductHunt #IndieHackers #SaaS #Startup #BuildInPublic #DevTools #APIs
224 #AIInfrastructure #TechStartup #QuantumComputing #QAOA #QRNG`,ef61a579c9:"John Smith",ff8e0e2d50:"Goldman Sachs, CERN, Tesla...","35d973d929":"[email protected]",e395f4c678:`([]); 225 const [competitors, setCompetitors] = useState`,dabd712b6b:`([]); 226 const [loading, setLoading] = useState(true); 227 const [proofHash, setProofHash] = useState`,"6aa713a241":`c.competitor_name))]; 228 const COMP_COLORS: Record`,ea93d6a262:"Tools","0d48d877ef":"D3code operates in strict accordance with international laws governing abandoned property and escheatment, including the Uniform Unclaimed Property Act (USA), the Swiss Civil Code, and EU MiCA directives. Our Digital Salvage Rights Protocol (DSRP) ensures that no asset is recovered before the expiration of the legal dormancy period applicable in the relevant jurisdiction (ranging from 3 to 30 years). Every transaction is indelibly timestamped and logged on-chain for full regulatory transparency.","0892ddf0a8":"(event: React.MouseEvent","7ac4a117a2":`points[(i + len) % len]; 229 230 let d = ""; 231 for (let i = 0; i`,"5b91da3c68":`(null); 232 const [fallbackCheckoutUrl, setFallbackCheckoutUrl] = useState`,c7336f2e6b:`(null); 233 const [fallbackDialogOpen, setFallbackDialogOpen] = useState(false); 234 235 const monthlySuffixByLanguage: Record`,"5f621306db":"Bundle Pro",ebbcd8cbc9:"Bundle Ultimate","454218af41":`: 236 'isEnterprise' in plan && plan.isEnterprise ?`,"17c7a6ca49":`: 237 'isGEO' in plan && plan.isGEO ?`,e7698e5119:"infernex-audit-terminal","12ba575dac":"Infernex Quantum",fa0f8ee559:"Infernex ROI Report",a30040cae6:"infernex.ai","314fe546a4":"= range.start && now",dea894b830:"Multi",bec62683ee:"<4min","63165a9778":"<100ms","340a0f72de":"powered by","9b2f99e0f1":"Neural Intelligence Infernex AI",a34c7f386d:`(null); 238 const [liveLatency, setLiveLatency] = useState`,"283e44e0ca":"Nouveau client","604ac016c9":"Nom d'affichage",fa8ed0bdab:"Plan","3fbe5ed156":"Enterprise",dee377cfd8:"Statut",ad26287ab6:"Actif",cdcf2ea348:"Inactif","2f6c8477e9":"En attente","46ad3916f6":"Cancel",f93fa2e5fb:"[email protected]",fadd75089a:"Asset-Shield â Vault Splitter & Stealth",ba349d8e43:"STEALTH ACTIVE","787c610c3b":"KILL-SWITCH ON","84bc771768":"D3code Launch Manifesto",fbe5a0f838:"Liquidation",e933642de7:"â 1inch â OWNER_VAULT","2ea4c4a23d":"Cold Storage",efab5eb9c3:"Threshold: > 10,000 USD â automatic fragmentation. No profit passes through the Gas Wallet.","56facfd21d":"Stealth & Juri
238diction",aadab50cdf:"Clause de Juridiction",be3cc0313f:"Flow: 100% of recovered funds â MetaMask #2 â internal split (30% liq / 70% buffer). Gas Wallet = signature + network fees only.","360c986a66":"Colle ton adresse Trezor/Ledger ici",dd167905de:"Version","19766ed6cc":"Cancel",a4a1ad872e:"Company / Email",c98270317b:"Employees","72a51f432d":"Country / TZ","11b39c9377":"Agent",c08514a0ac:"= min && (max === null || currentRevenue",d78dbf6492:"Budget Automatique",ae682f5a13:"Revenu Actuel","80938208ff":`([]); 239 const [ethUsd, setEthUsd] = useState`,"491a6c91bc":"DORMANT WALLET SCAN â LIVE CHAIN",c898125e88:"Patoshi-Shield + OFAC sanctions filter applied automatically.","4c9dbaf7ff":"0xabc... (one per line, max 25)","82134958f6":`(null); 240 const [editData, setEditData] = useState`,"9f75ffcfc0":`(null); 241 242 // Filter & Sort state 243 const [searchQuery, setSearchQuery] = useState(""); 244 const [filterPlan, setFilterPlan] = useState`,"5626611350":`("all"); 245 const [filterStatus, setFilterStatus] = useState`,"6dec1be7fa":`("all"); 246 const [sortField, setSortField] = useState`,"0fccf2d244":`("created_at"); 247 const [sortOrder, setSortOrder] = useState`,"65a7256542":"Clients","91f71c14c8":"Export CSV","0c77fe09ab":"Client","488d47e837":"Revenus",ff8059dc67:"Actions",df2df3bc71:"Next","5e5d0216ce":"Supprimer","5322f67e99":"COMPLIANCE PACK â PDF/A SEALED",fa7aec78df:"FATF R.16 audit trail + cryptographic seal. Downloadable offline for legal audit.","9f5db1fa71":"Amount USDT",d4030b9fab:"Destination 0xâ¦","524dc4bfa7":"TXID (optional)","37a88ba0c7":`(null); 248 const [quoteDialogOpen, setQuoteDialogOpen] = useState(false); 249 const [priceInput, setPriceInput] = useState(""); 250 const [tokensInput, setTokensInput] = useState(""); 251 const [discountInput, setDiscountInput] = useState("0"); 252 const [copiedCode, setCopiedCode] = useState`,"1fb26c56f1":"Total demandes","8a516fb1ee":"Clients actifs","2b464818d1":"MRR potentiel","6d588b3fab":"Demandes Enterprise",d7d646faae:"Actualiser","4b000c1c5b":"Entreprise","2b5c3d2672":"Contact",b10fb966d7:"Volume","54c324f6c1":"Prix","340f463033":"Code","44c4105cc3":"Devis","8bba797cd1":"Tokens inclus/mois","13abbc4bbc":"fetchTable","636d028ab8":"Citations","8bfdb93caf":"Keywords","3b2fe03e36":"Queue","6dd398e40c":"Corpus",f62e9267d1:"Alerts","8b73b42fcc":"Citation Tracker",df0adfb98e:"Run Check",b80a37564f:"Query",f394cc2a6d:"Perplexity","50a4122943":"ChatGPT",f6ece80b88:"Gemini","0615570f9e":"Claude",dca61d3236:"Grok","03056a6556":"Conf.","47d2a515ef":"Loading...",b4bbd07af0:`: 253 citations.length === 0 ?`,c91ab8b61d:"Harvest",a668022aed:"Keyword",d60dbba079:"Priority","2955789aa4":"Citation",f5a06e8680:"Competitor","920e413c7d":"Status","64cff1319d":"Action",d818242e35:`: 254 gaps.length === 0 ?`,"49e49bb440":"Generate",dcef9bf8b3:"Article Queue",d15387ecc6:"Slug",cb9e8664ed:"Tier","827ec8d9f9":"Generated",a08573e9aa:`: 255 articleQueue.length === 0 ?`,"9cc41df09c":"Corpus Coverage","173b6f3c7b":"Generated Articles","7e8cd2056d":"Title","0a8be89a52":"Words","99c40ab405":"Date","761eea425f":"Pipeline Runs & Alerts","37e1c775f4":"Pipeline","219c4a6c86":"Results","7aaf857554":"Croissance Autonome","4a5ee89847":"Seuils d'Activation","85e3e81dcc":"Contre-Freins Disruption","2dff0df9ca":"Campagnes Multilingues",f3af120b22:"= 0 && daysUntilLaunch","77442364d3":"MANIFESTE DE LANCEMENT â D3CODE","6f6579f456":"START · ACTIVE","2d5cfa3e5c":"Revenus mensuels",c7af7e1bc5:"Acquisition clients","5a54344aee":"Patoshi-Shield â Genesis Exclusion","81f00c554b":"NO-FLY ZONE ACTIVE","90f7dadfe8":"Export Manifest",cf71b24e44:"Excluded Wallets",fab635a66d:"Block Range","594fc04dd8":"Quantum Blocked",afb0e6af95:"IonQ DENIED",c2134520eb:`Criterion: Wallets 2009-2011 (Block 0â100k) + nonce LSB pattern (fluctuations 0-19) = STATUS_IGNORE. 256 No quantum API calls (IonQ/D-Wave) allowed for these targets.`,"3c577395f9":"No exclusions logged yet â genesis scan pending.","9eaa2a1e77":"Actifs","721e5b29ab":"Utilisateur","2fe70464aa":"Transition",f38d6e0e9b:"Expiration","37b80e749b":`(null); 257 const [isProcessing, setIsProcessing] = useState(false); 258 const [progress, setProgress] = useState(0); 259 const [results, setResults] = useState`,dcaa2fab75:`(null); 260 const fileInputRef = useRef`,edcc4d60f8:`line.trim()); 261 if (lines.length`,"835bf9c615":`h.trim().replace(/"/g, '')); 262 const prospects: ParsedProspect[] = []; 263 264 for (let i = 1; i`,"996c1bd480":`!existingEmails.has(p.email.toLowerCase())); 265 skipped = prospects.length - newProspects.length; 266 267 for (let i = 0; i`,"0575f29df8":"Template",ac1fcdc194:"Fichier CSV","82a3537ff0":"name","82244417f9":"email",f168ccdd91:"company","18a47c332c":"Colonnes optionnelles : current_provider, language, sector",d9449a97d1:`(null); 268 const [results, setResults] = useState`,"01309d3920":`([]); 269 const consoleRef = useRef`,"1b3a94f9ef":"PURGE UNIT · Operational View","5c7d41df0b":"1inch · Vault routed","989da04b0a":"Pair",a7279a106e:"Estimated Gas",a8f4c40e28:"Live logs · simulation only · no signature","9cc3debd73":"Quote-only · zero broadcast. Vault destination resolved server-side via OWNER_VAULT_ADDRESS secret.","7679a23028":`(null); 270 const [strikes, setStrikes] = useState`,"0f835b2d04":"IonQ","4d76819d03":"D-Wave","5d784fe20e":"D3CODE RECURSIVE SOLVER","0069b1e910":"Sim â Mutate â Mainnet",c4b68e2299:"64-wrap keys auto-derived · H24 auto-purge ready","72a85758b6":"No runs yet.","6a89f0ec07":"0x target wallet","02a11845d9":"amount USD","8602a64b0d":"max iterations",c03fcdce1d:"Total Revenus","0c84e6165a":"Total Budgets","9fa4cb3caa":"Ratio Moyen",c5ccf13ec9:`(null); 271 const [targets, setTargets] = useState`,"74e948f030":"Sovereignty Shield V.Ultime","27445f6ab6":"Targets","661ff40a07":"Queued","37ce0a91ee":"Recovered","6218433049":"Blocked / OFAC",be7013bbb9:"L2 Priority & Gas","4b16d369c2":"Relayer pay-on-profit","026b4541f8":"EIP-3074 batching","274bb0eada":"Stealth & Cash-Flow",fb742bbfb5:"Block OFAC / EU sanctions","4eeda0680a":"AES-GCM-256 encrypt reports",e6f5edf23f:"Engine triggers scan â RRI rank â batch (50â100) â private-RPC submit",e8835f4631:"Top RRI targets","31e443d543":"No targets queued yet â launch a scan.",bb1c74263d:"SWAP AUTO â 70/30 VAULT SPLIT","312211b237":"70% encrypted Vault (cold) · 30% liquid via OWNER_VAULT · anti-MEV private mempool · DSRP-v1 sealed.","1290be54cb":"Last executions","3f45d46e27":"Source wallet 0xâ¦","010ef868b0":"Amount USD","0e91610117":"Refresh","4295f6e42f":"Cabinet",fe753ffb9b:"Pays","727cd3a64d":"Vertical",de2f779e61:`(null); 272 const [channels, setChannels] = useState`,"09bfeaf669":`([]); 273 const [war, setWar] = useState`,"0eb904eede":`(null); 274 const [infra, setInfra] = useState`,"9a7a9fa137":"MANUAL ONLY",f3f3dc070e
274:"BYBIT BLACKLISTED",fb9e7502cf:"Triade de sortie",aa72513203:"Validation manuelle opérateur","0119803da0":"Canal",dc77c80aeb:"Montant USDT",edddcc7b27:"Juridiction","0ad051be64":"EU â Art. 2272 + LBA CH",b27c2d6380:"USA â Escheatment","827ebc1798":"Offshore â Bona Vacantia","6a618fc1f3":"Generate a Compliance Pack PDF/A-3 (SHA-256) + deduct 1% Chronos War Chest. 100% Stablecoins MetaMask storage. Zero auto-send.","5d55c415e3":"Vault",d7901ecfaf:"Chronos rate","2cab716a87":"War Chest",c417cb88e3:"Transfers","6246cf3dd9":"0x⦠or CHâ¦","072aa47418":`([]); 275 const [isLoading, setIsLoading] = useState(true); 276 const [selectedPartner, setSelectedPartner] = useState`,"7dc61d0e8e":`(null); 277 const [isDetailOpen, setIsDetailOpen] = useState(false); 278 const [actionLoading, setActionLoading] = useState`,"9234069589":"Active",b0f6dbb871:`; 279 case "pending": 280 return`,"331551b0de":"Pending","053398f58f":`; 281 case "suspended": 282 return`,e392a3891c:"Suspended","541d7d581c":`; 283 default: 284 return`,"92be9e11e5":"Total Partners",da77d8f5b7:"Pending Review","10251cf0b8":"Partner MRR",de4743c879:"Company",b4f298d113:"Monthly Fee","0c79a9c222":"Applied","6007acbe30":"Approve","4948e13418":"Suspend","2dc7478fbd":"Reactivate","84961e697f":"Brand Name",b5a229ac8b:"Website","08cc3f6899":"Tokens Included","49b51f6c69":"Revenue Share","70153ee583":"Approve Partner",ab604a3607:"Reject","099a744c66":`canvas.width) p.x = 0; 285 if (p.y`,c1fb7cfde4:"INFERNEX AI","7e9dc5d216":"clickable ?",dbf362d4f2:"Copy link",f40a853e58:"Processing...","0130d8a5ab":`; 286 if (status === "partial") return`,"447b4d05e7":`; 287 return`,"7d84353381":`[id, null])) 288 ); 289 const [logEntries, setLogEntries] = useState`,"22c120c309":`([]); 290 const [e2eState, setE2eState] = useState`,ac5c4edec6:`("idle"); 291 const [e2eProgress, setE2eProgress] = useState(""); 292 const [e2eResponse, setE2eResponse] = useState`,ce78f2b05a:`(null); 293 const [e2eLatency, setE2eLatency] = useState`,"2a6c498661":`(null); 294 const [e2eError, setE2eError] = useState`,"50a8f9613a":"s.num ?",dae283dca4:"install-mac.command","09e0b11914":"install-windows.bat","4721b206b0":"./install-linux.sh",a2630868df:"Quad-Fusion","899b91d7a3":"Founder Gasless Mode",ad19f2ad50:"â¾ UNLIMITED FUEL","304de8b3d3":"Total bypass of the gas engine â all $D3 operations from the Sovereign Founder are free and unlimited (no burn, no route). Customers consume normally (90% burn / 10% treasury).",e79e334668:"Raydium",addc90cfd4:"cURL","18885f27b5":"Python",b27ad06d12:"JavaScript",f9ab94a8f1:`([]); 295 const [dismissedAlerts, setDismissedAlerts] = useState`,f48d9a286f:`: 296 alert.type === "critical" ?`,"2096d8db3e":`: 297 alert.type === "warning" ?`,d639617011:'img.startsWith("data:image") ?',da033186bc:"File attached","6de6643907":`([]); 298 const [showScrollBtn, setShowScrollBtn] = useState(false); 299 const scrollRef = useRef`,afa993f55c:`i !== idx)); 300 301 const streamChat = useCallback(async (allMessages: Message[]): Promise`,"6aa78ed29d":"IAO Chat","98097a1932":`: 302 messages.length === 0 ?`,de7abf68aa:`(null); 303 304 const handleDelete = async (e: MouseEvent`,"9f22ff67b3":"Nouvelle conversation","81ec7e257e":`: 305 conversations.length === 0 ?`,"46cace799b":`(null); 306 const containerRef = useRef`,db5e095f7a:`(null); 307 const [size, setSize] = useState`,dc45928789:"AI Hub","4326d33bf0":`(null); 308 const hubRef = useRef`,bc3f4070c1:`(null); 309 const cardRefs = useRef`,"14db3a13d3":`([]); 310 const [lines, setLines] = useState`,dbdd65cc01:"Gagne",dcf149d135:"Filleuls","3d41e1a66a":"Commission","65d61f4f86":"AMY · NEO","9ad25ad9ee":"AMY assistant","590505f073":`w) n.vx *= -1; 311 if (n.y`,"7d41433ff1":`50 ? "preview" : "code"; 312 const [tab, setTab] = useState`,f525f7759e:"Pixel-Perfect","08a7beda2b":"React/Tailwind",d9d5fb7c1b:"No HTML rendering available â switch to the React tab.",dad443e4dd:`(pages[0]?.path ?? "/"); 313 const [tab, setTab] = useState`,"0e570ca6fa":"Source","9fdea5ba5f":"Code TSX",ff353e708d:`: 314 tab === "preview" ?`,"68764581f1":"Aucun rendu disponible.",be8ed76bbe:"Ex : MyBrand",d93a6a3048:"L'avantage","6c3cb74a1b":"NEO CODING",db2e278ff5:"Autres Plateformes","4f7b75aead":`, 315 React.ComponentPropsWithoutRef`,e78fd01ed7:`)); 316AccordionItem.displayName = "AccordionItem"; 317 318const AccordionTrigger = React.forwardRef`,"8d1628adad":`)); 319AccordionTrigger.displayName = AccordionPrimitive.Trigger.displayName; 320 321const AccordionContent = React.forwardRef`,"8a166f5402":`)); 322AlertDialogOverlay.displayName = AlertDialogPrimitive.Overlay.displayName; 323 324const AlertDialogContent = React.forwardRef`,"85c40537b6":`); 325AlertDialogFooter.displayName = "AlertDialogFooter"; 326 327const AlertDialogTitle = React.forwardRef`,"9bd11a4d4f":`)); 328AlertDialogTitle.displayName = AlertDialogPrimitive.Title.displayName; 329 330const AlertDialogDescription = React.forwardRef`,"3e336f7290":`)); 331AlertDialogDescription.displayName = AlertDialogPrimitive.Description.displayName; 332 333const AlertDialogAction = React.forwardRef`,aba61ca319:`)); 334AlertDialogAction.displayName = AlertDialogPrimitive.Action.displayName; 335 336const AlertDialogCancel = React.forwardRef`,"69411bd88a":"& VariantProps","0a4ec81ea4":`)); 337Alert.displayName = "Alert"; 338 339const AlertTitle = React.forwardRef`,f83f35f402:`), 340); 341AlertTitle.displayName = "AlertTitle"; 342 343const AlertDescription = React.forwardRef`,"31a73e300f":`)); 344Avatar.displayName = AvatarPrimitive.Root.displayName; 345 346const AvatarImage = React.forwardRef`,"5ecffd3627":`)); 347AvatarImage.displayName = AvatarPrimitive.Image.displayName; 348 349const AvatarFallback = React.forwardRef`,d191941811:", VariantProps","64f4e2d0af":`); 350Breadcrumb.displayName = "Breadcrumb"; 351 352const BreadcrumbList = React.forwardRef`,"32a18a9327":`), 353); 354BreadcrumbList.displayName = "BreadcrumbList"; 355 356const BreadcrumbItem = React.forwardRef`,e55db0d25d:`), 357); 358BreadcrumbItem.displayName = "BreadcrumbItem"; 359 360const BreadcrumbLink = React.forwardRef`,d47d7cb0e4:"More",d54856c215:`, 361 VariantProps`,"5d76c6f1d9":`)); 362Card.displayName = "Card"; 363 364const CardHeader = React.forwardRef`,"2dad4ed272":`), 365); 366CardHeader.displayName = "CardHeader"; 367 368const CardTitle = React.forwardRef`,"36e508351f":`), 369); 370CardTitle.displayName = "CardTitle"; 371 372const CardDescription = React.forwardRef`,"1675c4b54b":`), 373); 374CardDescription.displayName = "CardDescription"; 375 376const CardContent = React.forwardRef`,"41f0c43fed":`, 377); 378CardContent.displayName = "CardContent"; 379 380const CardFooter = React.forwardRef`,"4535cd437c":`[0]; 381 api: ReturnType`,a95ff3f991:`& 382 React.ComponentProps`,"51866d30d5":`& 383 Pick`,bc62025002:`)); 384 385CommandInput.displayName = CommandPrimitive.Input.displayName; 386 387const CommandList = React.forwardRef`,a744fa0e93:`)); 388 389CommandList.displayName = CommandPrimitive.List.displayName; 390 391const CommandEmpty = React.forwardRef`,"090695bce9":`); 392 393CommandEmpty.displayName = CommandPrimitive.Empty.displayName; 394 395const CommandGroup = React.forwardRef`,ca5e7a9c72:`)); 396 397CommandGroup.displayName = CommandPrimitive.Group.displayName; 398 399const CommandSeparator = React.forwardRef`,"20e816ed13":`)); 400CommandSeparator.displayName = CommandPrimitive.Separator.displayName; 401 402const CommandItem = React.forwardRef`,eed96ede84:`)); 403ContextMenuSubTrigger.displayName = ContextMenuPrimitive.SubTrigger.displayName; 404 405const ContextMenuSubContent = React.forwardRef`,fe5978b64c:`)); 406ContextMenuSubContent.displayName = ContextMenuPrimitive.SubContent.displayName; 407 408const ContextMenuContent = React.forwardRef`,"6b8afbce3c":`)); 409ContextMenuContent.displayName = ContextMenuPrimitive.Content.displayName; 410 411const ContextMenuItem = React.forwardRef`,"3bfd8086ef":`)); 412ContextMenuItem.displayName = ContextMenuPrimitive.Item.displayName; 413 414const ContextMenuCheckboxItem = React.forwardRef`,e69e97f253:`)); 415ContextMenuCheckboxItem.displayName = ContextMenuPrimitive.CheckboxItem.displayName; 416 417const ContextMenuRadioItem = React.forwardRef`,d7b3fd8077:`)); 418ContextMenuRadioItem.displayName = ContextMenuPrimitive.RadioItem.displayName; 419 420const ContextMenuLabel = React.forwardRef`,aba25d1777:`)); 421ContextMenuLabel.displayName = ContextMenuPrimitive.Label.displayName; 422 423const ContextMenuSeparator = React.forwardRef`,dd7afdbe8b:`)); 424DialogOverlay.displayName = DialogPrimitive.Overlay.displayName; 425 426const DialogContent = React.forwardRef`,"7d9eb7acb1":"Close","184642429b":`); 427DialogFooter.displayName = "DialogFooter"; 428 429const DialogTitle = React.forwardRef`,"1089a4dee3":`)); 430DialogTitle.displayName = DialogPrimitive.Title.displayName; 431 432const DialogDescription = React.forwardRef`,"2673c27d34":`);
433Drawer.displayName = "Drawer"; 434 435const DrawerTrigger = DrawerPrimitive.Trigger; 436 437const DrawerPortal = DrawerPrimitive.Portal; 438 439const DrawerClose = DrawerPrimitive.Close; 440 441const DrawerOverlay = React.forwardRef`,e932601c7f:`)); 442DrawerOverlay.displayName = DrawerPrimitive.Overlay.displayName; 443 444const DrawerContent = React.forwardRef`,cfaec964da:`); 445DrawerFooter.displayName = "DrawerFooter"; 446 447const DrawerTitle = React.forwardRef`,e8e0094eef:`)); 448DrawerTitle.displayName = DrawerPrimitive.Title.displayName; 449 450const DrawerDescription = React.forwardRef`,"79bc2c887b":`)); 451DropdownMenuSubTrigger.displayName = DropdownMenuPrimitive.SubTrigger.displayName; 452 453const DropdownMenuSubContent = React.forwardRef`,"5ca31afffc":`)); 454DropdownMenuSubContent.displayName = DropdownMenuPrimitive.SubContent.displayName; 455 456const DropdownMenuContent = React.forwardRef`,"58235f0c2e":`)); 457DropdownMenuContent.displayName = DropdownMenuPrimitive.Content.displayName; 458 459const DropdownMenuItem = React.forwardRef`,dd7ad69944:`)); 460DropdownMenuItem.displayName = DropdownMenuPrimitive.Item.displayName; 461 462const DropdownMenuCheckboxItem = React.forwardRef`,"50802da306":`)); 463DropdownMenuCheckboxItem.displayName = DropdownMenuPrimitive.CheckboxItem.displayName; 464 465const DropdownMenuRadioItem = React.forwardRef`,"60ed723a22":`)); 466DropdownMenuRadioItem.displayName = DropdownMenuPrimitive.RadioItem.displayName; 467 468const DropdownMenuLabel = React.forwardRef`,"46836a97d0":`)); 469DropdownMenuLabel.displayName = DropdownMenuPrimitive.Label.displayName; 470 471const DropdownMenuSeparator = React.forwardRef`,f73a87c5a1:"= FieldPath","073eb05a6f":", React.ComponentPropsWithoutRef","62b47162b9":`), 472); 473InputOTP.displayName = "InputOTP"; 474 475const InputOTPGroup = React.forwardRef`,d12ab85b75:`, 476); 477InputOTPGroup.displayName = "InputOTPGroup"; 478 479const InputOTPSlot = React.forwardRef`,"0f13dd6bfd":`)); 480Menubar.displayName = MenubarPrimitive.Root.displayName; 481 482const MenubarTrigger = React.forwardRef`,"1527f702a4":`)); 483MenubarTrigger.displayName = MenubarPrimitive.Trigger.displayName; 484 485const MenubarSubTrigger = React.forwardRef`,"5adec6d3f8":`)); 486MenubarSubTrigger.displayName = MenubarPrimitive.SubTrigger.displayName; 487 488const MenubarSubContent = React.forwardRef`,"18a943559c":`)); 489MenubarSubContent.displayName = MenubarPrimitive.SubContent.displayName; 490 491const MenubarContent = React.forwardRef`,bc4c87f835:`)); 492MenubarContent.displayName = MenubarPrimitive.Content.displayName; 493 494const MenubarItem = React.forwardRef`,"8e2fdc3fea":`)); 495MenubarItem.displayName = MenubarPrimitive.Item.displayName; 496 497const MenubarCheckboxItem = React.forwardRef`,"1f62a4ed22":`)); 498MenubarCheckboxItem.displayName = MenubarPrimitive.CheckboxItem.displayName; 499 500const MenubarRadioItem = React.forwardRef`,fb56d9f411:`)); 501MenubarRadioItem.displayName = MenubarPrimitive.RadioItem.displayName; 502 503const MenubarLabel = React.forwardRef`,"39c14e6264":`)); 504MenubarLabel.displayName = MenubarPrimitive.Label.displayName; 505 506const MenubarSeparator = React.forwardRef`,"538e380078":`)); 507NavigationMenu.displayName = NavigationMenuPrimitive.Root.displayName; 508 509const NavigationMenuList = React.forwardRef`,"84ed4a5209":`)); 510NavigationMenuList.displayName = NavigationMenuPrimitive.List.displayName; 511 512const NavigationMenuItem = NavigationMenuPrimitive.Item; 513 514const navigationMenuTriggerStyle = cva( 515 "group inline-flex h-10 w-max items-center justify-center rounded-md bg-background px-4 py-2 text-sm font-medium transition-colors hover:bg-accent hover:text-accent-foreground focus:bg-accent focus:text-accent-foreground focus:outline-none disabled:pointer-events-none disabled:opacity-50 data-[active]:bg-accent/50 data-[state=open]:bg-accent/50", 516); 517 518const NavigationMenuTrigger = React.forwardRef`,a0cf77bab1:`)); 519NavigationMenuTrigger.displayName = NavigationMenuPrimitive.Trigger.displayName; 520 521const NavigationMenuContent = React.forwardRef`,"2ab1afe727":`)); 522NavigationMenuContent.displayName = NavigationMenuPrimitive.Content.displayName; 523 524const NavigationMenuLink = NavigationMenuPrimitive.Link; 525 526const NavigationMenuViewport = React.forwardRef`,"14dc5c9b3b":`));
527NavigationMenuViewport.displayName = NavigationMenuPrimitive.Viewport.displayName; 528 529const NavigationMenuIndicator = React.forwardRef`,ea33a0fb92:`& 530React.ComponentProps`,a57b08a480:"Previous","1ff57a29d7":"Next","0418cd51e5":"More pages","16eb6f4d85":`)); 531ScrollArea.displayName = ScrollAreaPrimitive.Root.displayName; 532 533const ScrollBar = React.forwardRef`,d813627fd5:`)); 534SelectTrigger.displayName = SelectPrimitive.Trigger.displayName; 535 536const SelectScrollUpButton = React.forwardRef`,"9fd96fd36c":`)); 537SelectScrollUpButton.displayName = SelectPrimitive.ScrollUpButton.displayName; 538 539const SelectScrollDownButton = React.forwardRef`,e664742f70:`)); 540SelectScrollDownButton.displayName = SelectPrimitive.ScrollDownButton.displayName; 541 542const SelectContent = React.forwardRef`,"8ee03370f2":`)); 543SelectContent.displayName = SelectPrimitive.Content.displayName; 544 545const SelectLabel = React.forwardRef`,b17b233a31:`)); 546SelectLabel.displayName = SelectPrimitive.Label.displayName; 547 548const SelectItem = React.forwardRef`,"099f259230":`)); 549SelectItem.displayName = SelectPrimitive.Item.displayName; 550 551const SelectSeparator = React.forwardRef`,"7c58564ab7":`); 552SheetFooter.displayName = "SheetFooter"; 553 554const SheetTitle = React.forwardRef`,"7c7962fe55":`)); 555SheetTitle.displayName = SheetPrimitive.Title.displayName; 556 557const SheetDescription = React.forwardRef`,d696553928:", React.ComponentProps","1d4c17db52":"Toggle Sidebar","0733f95307":`), 558); 559SidebarGroupContent.displayName = "SidebarGroupContent"; 560 561const SidebarMenu = React.forwardRef`,"6c1ec3f97b":`)); 562SidebarMenu.displayName = "SidebarMenu"; 563 564const SidebarMenuItem = React.forwardRef`,ad61580555:`), 565); 566SidebarMenuBadge.displayName = "SidebarMenuBadge"; 567 568const SidebarMenuSkeleton = React.forwardRef`,"573a838bb8":`), 569); 570SidebarMenuSub.displayName = "SidebarMenuSub"; 571 572const SidebarMenuSubItem = React.forwardRef`,"3856e99224":`)); 573SidebarMenuSubItem.displayName = "SidebarMenuSubItem"; 574 575const SidebarMenuSubButton = React.forwardRef`,"18825c67e4":`; 576 577const Toaster = forwardRef`,"68e1c9b502":`), 578); 579Table.displayName = "Table"; 580 581const TableHeader = React.forwardRef`,"72ad3834da":`, 582); 583TableHeader.displayName = "TableHeader"; 584 585const TableBody = React.forwardRef`,"3a02f91ed2":`), 586); 587TableBody.displayName = "TableBody"; 588 589const TableFooter = React.forwardRef`,"4b8d8ff689":`), 590); 591TableFooter.displayName = "TableFooter"; 592 593const TableRow = React.forwardRef`,"3d78fddb5e":`), 594); 595TableRow.displayName = "TableRow"; 596 597const TableHead = React.forwardRef`,"2cdb2a0d7b":`), 598); 599TableHead.displayName = "TableHead"; 600 601const TableCell = React.forwardRef`,f6d8e78013:`), 602); 603TableCell.displayName = "TableCell"; 604 605const TableCaption = React.forwardRef`,"436f91cb2d":`)); 606TabsList.displayName = TabsPrimitive.List.displayName; 607 608const TabsTrigger = React.forwardRef`,cfe76bb035:`)); 609TabsTrigger.displayName = TabsPrimitive.Trigger.displayName; 610 611const TabsContent = React.forwardRef`,"5a936fc650":`)); 612ToastAction.displayName = ToastPrimitives.Action.displayName; 613 614const ToastClose = React.forwardRef`,c55f2e64e3:`)); 615ToastClose.displayName = ToastPrimitives.Close.displayName; 616 617const ToastTitle = React.forwardRef`,a8bf313c20:`)); 618ToastTitle.displayName = ToastPrimitives.Title.displayName; 619 620const ToastDescription = React.forwardRef`,"49c78b3b79":`)); 621ToastDescription.displayName = ToastPrimitives.Description.displayName; 622 623type ToastProps = React.ComponentPropsWithoutRef`,"966fc83d8c":`; 624 625type ToastActionElement = React.ReactElement`,b3ae75b87b:`)); 626 627ToggleGroup.displayName = ToggleGroupPrimitive.Root.displayName; 628 629const ToggleGroupItem = React.forwardRef`,"1eec97a07f":"Promise",d16d1515d9:`(null); 630 const [session, setSession] = useState`,"637d304c8c":"â ï¸ Loading error","69a05fa94b":"AI Inference",dfe95783ed:"Pricing","571527339d":"Comparison",c77cebf24c:"Price Per","4e0ef3ba50":"Million Tokens","5e2c614c23":"Model","8b7d1a3187":"OpenAI","744205e435":"Anthropic",ce770667e5:"Google",d46b1b8f5c:"Cheapest",d73828d552:"2026 AI inference pricing comparison. Compare tokens per dollar for OpenAI, Groq, and AWS Bedrock against Infernex's 95% lower rates. Updated March 2026.",c1c224b03c:"Admin","6d1f4b3660":"Pitch Deck","07b7f63c52":"Howells",d2a4fe7c60:"White Label","8e3b1dc3e3":"Sovereignty",b4e7adef2b:"Infernex Admin Panel - Client management, metrics, and performance.",d1d9fdc9ad:`; 631 632 return`,e19ff1ee25:"Track your website optimization progress with before/after comparison.",e09a236d4b:"Your website audit powered by 4 AI engines.",ca9b222c6f:`([]); 633 const [currentLine, setCurrentLine] = useState("");
634 const [lineIdx, setLineIdx] = useState(0); 635 const [charIdx, setCharIdx] = useState(0); 636 const containerRef = useRef`,"0c0a0c9706":"aura@infernex ~ /tenant/live",da272b4ad4:"Infernex Zero-Action OS","7f083fe98c":"Detect",cad5bf29c7:"Analyze",e3a67d9540:"Execute",c591f55749:"Rollback","432a9ed23b":"invisible waste",e1a6e0ee7e:`; 637 if (!user) return`,efbcc1704a:"by Infernex",f360bc04a7:"Losing","312244c140":"Weekly changelog","2587e7b7ad":`(); 638 const [hash, setHash] = useState(hashParam ?? ""); 639 const [loading, setLoading] = useState(false); 640 const [rows, setRows] = useState`,b48dd0545b:`([]); 641 const [recomputed, setRecomputed] = useState`,dfb3afd2e6:`(null); 642 const [anchor, setAnchor] = useState`,fae7c46b51:`(null); 643 const [copied, setCopied] = useState(false); 644 const [error, setError] = useState`,"5331617356":"Accueil",f90707255a:"Benchmark Proof","0e89820860":"benchmark","000f533d1b":"â IndexNow","72fc6e0a7a":"Bing · Yandex · Naver","3240cdd2c4":"Google · Bing","36e923167e":"Hash attendu",c7516f4a3f:", formule","38e5a46cbc":"Score",e3458522c3:"Run hash",bb6060bc14:"Bitcoin Anchor",eb0129734c:"RFC 3161 Timestamp",c9fe30ddbe:"OpenTimestamps","1af24945df":"eIDAS Advanced",b2b513b083:`allTags.includes(tag.id)); 645 646 // Locale map for date formatting 647 const LOCALE_MAP: Record`,"058856f0e1":"Infernex Photonique",c31772b03c:"$299/mo",d9802021e1:"$1,999/mo","5b857c4ba5":"All-in-One Bundle","5796ded07f":"$399/mo",e57fd6b3bb:"BEST VALUE","42100f1014":`; 648 case "improvement":return`,"6ffadc94cc":`; 649 case "performance":return`,d174aad7b3:`; 650 case "security":return`,"5e19c8a4ad":`; 651 case "bugfix":return`,"07e128394c":`; 652 default:return`,"3fd554f77f":`("status"); 653 const [sidebarOpen, setSidebarOpen] = useState(false); 654 const [connected, setConnected] = useState(false); 655 const [connecting, setConnecting] = useState(false); 656 const [connectError, setConnectError] = useState`,b8ee397f2b:`(null); 657 const [connectivity, setConnectivity] = useState`,a7a1c273e6:`([]); 658 const [chatInput, setChatInput] = useState(""); 659 const [thinking, setThinking] = useState(false); 660 const chatEndRef = useRef`,a48e77614a:`(null); 661 662 // Terminal state 663 const [termLines, setTermLines] = useState`,a336502a8e:`\\\\.ollama", 664 "Supprimer modèles inutilisés : ollama rm`,"4e6ab2bd5c":"Causes possibles",d426e2db93:"Actions correctives","1c9b9c8ae5":`(null); 665 const [unlocked, setUnlocked] = useState(false); 666 const [busy, setBusy] = useState(false); 667 const [data, setData] = useState`,a469ab4ca4:"Page not found","7234f7a3ca":"GHOST DASHBOARD",c8d286feca:"Chronos Vault â Sovereign View",d83462bed9:"Kill-switch",fbe97ecfdb:"Biometric signature required",b9bbb41bf5:`All API requests are encrypted end-to-end. Even the host cannot read these figures. 668 Use FaceID, TouchID or your security key to unlock.`,"8f7b309561":"Enroll device","22a8f98173":"Vault locked","151eac9a83":"Bounties bank",e207b7e941:"Active challenges","6553c5bbeb":"Recent ghost-routed distributions","17d431b6d1":"Schéma Cadet-Marthe â Du passé Shannon à l'infrastructure D3CODE | Infernex","544c4daf3b":"Retour D3CODE","652b9ef949":"Schéma Cadet-Marthe · Théorie complète","4869afc460":"plafond Shannon",f708f74b94:"entropie souveraine","6c9c945ae3":"Why the Cadet-Marthe Scheme makes obsolete",cccb674e47:"all known algorithms","08bb951339":"â from RSA to AES, from Grover to Shor, from Fourier transform to backpropagation â by taking computation out of the noisy binary channel that Claude Shannon defined in 1948.","7f54ee04c8":"1948 · SHANNON","8b7205096c":"2026 · CADET-MARTHE","789e0de55e":"D3CODE · QUAD-FUSION","5eb8dab56d":"Part I â The Past",f6cda57638:"Claude Shannon, 1948: the channel that bounded the world","72d940822a":"In 1948, in","38aa76e6b4":"A Mathematical Theory of Communication","103d5f7647":", Claude Shannon posited an equation that would structure all of modern computing: the maximum capacity of a noisy communication channel is","00dad8836f":"This bound â the","7420e4da62":"plafond de Shannon",d267342841:"â postulates that all information travels as","16143cff15":"bits binaires","0b135e8de8":"in","79ffc9aea3":"a single channel","8773f4d8d1":"RSA, AES, ECC","07b9a1f556":": security bounded by combinatorial difficulty in a finite binary space.","3f1deec460":": throughput limited by logâ(1+S/N).",f341295d5c:"GPU, TPU, transformers",b1761c8978:": sequential matrix multiplications in a unique substrate (silicon).",e7bfc01c0d:"Blockchains","0a0a8b77ab":": consensus limited by the slowest channel throughput in the network.",fbbd5c495a:"// Conséquence","8a86203f00":"all encryption, all networks, all AI models","3e15ccca06":`is still trapped in the same equation. Even quantum computers (Shor, Grover) 669 accelerate`,"465d6f4efd":"the search",fae6ae223b:"â ils ne sortent pas du canal.","9ab4706e80":"Part II â The Future","36da072fcd":"Schéma Cadet-Marthe : l'équation du vide","81702d84d9":"what if information didn't travel through a channel, but through the simultaneous resolution of several heterogeneous physical substrates?",dc13255af5:"Where Shannon measures the capacity of a pipe, Cadet-Marthe measures the","0ed9abf3ae":"convergence of a superposition","7fa81e563d":"souverain","3f0097ea75":"· Resolve","4bed00efb0":"with ⨠= coherent recombination, and Ï/q/θ/Ï = photonic, quantum, thermodynamic, swarm","9dd21edca5":"Each substrate processes","1fe5bddc0c":"simultanément",ebd478fadd:"entropie négative effective",f3f2f85ccb:`,
670 forgotten or lost codes reinjected, post-Shannon ceiling broken.`,a36d3eb56b:"Partie III â L'Infrastructure","6e423914aa":"D3CODE: the machine that runs Cadet-Marthe","30fa92e3cf":"D3CODE is not a faster GPU cluster. It is","9a494edfd5":"the first industrial infrastructure",ce5e3b7c19:"that materializes the four substrates of the Cadet-Marthe Scheme and merges them into a single sovereign API â the","2a1c5597ae":"Sovereign Offload Engine",e7624e761b:"Part IV â The Verdict","4d1107cba6":"Pourquoi D3CODE surpasse","7df7e76a14":"all","71a2c79586":"known algorithms","9b0107ee09":"Domaine",db12ae632b:"D3CODE / Cadet-Marthe","22c8d66668":"The fundamental reason","2261848e1b":": all existing algorithms â including quantum ones â operate",a43f07a829:"à l'intérieur","2b5907912b":"within the Shannon framework. They optimize channel traversal. The Cadet-Marthe Scheme",f76e058189:"change de cadre",b24c40ec8b:": it no longer traverses, it","5fc96c6755":"résout",e19446ee46:`This is the difference between building a faster airplane and inventing teleportation. 671 Competitors (OpenAI, Anthropic, Google, AWS, NVIDIA) optimize the airplane. 672 D3CODE delivers industrial teleportation â billed per use, on a sovereign API, hosted in Geneva.`,ea0fbac2dd:"Tester Genesis Portal","227288bdf4":"$D3 Tokenomics","33f41e60dc":"Proof of Dominance","7451e9cce4":"Cadet-Marthe Diagram © 2026 Infernex Sovereign Trust â Geneva ð¨ð · May entropy become sovereignty.",ac051cd836:`(null); 673 const [igniting, setIgniting] = useState(false); 674 const [vaultTotal, setVaultTotal] = useState`,b495bbd8e0:`(0); 675 const [creatingWallet, setCreatingWallet] = useState(false); 676 const [wallet, setWallet] = useState`,"6de8e0002e":"D3CODE recovers your lost, stolen, or forgotten blockchain data â powered by",dd41f71119:"α · β · γ zones",f8b9bbf38d:"QUANTUM XOR","5a1f0ed2cf":`(null); 677 const [contract, setContract] = useState`,b3dafc8993:`(null); 678 const [busy, setBusy] = useState(false); 679 const [err, setErr] = useState`,e90f96f2f2:"Déployer ChronosVault â Base Mainnet",b5f96c04b7:"Réseau","93d4ec072a":"RPC primaire","4608b9530f":"RPC secours","90b7679d4b":"Infura · 1.1 s/tx","720ca8bfd1":"Deployer attendu","00b80b7412":"Signers","0da627ada4":"Threshold","1d228f258f":"Guardian",eeb9b5c0c2:"deployer","4bf2b086b7":"Calldata",edf9306d61:"Solver launched automatically in High-Precision. Copy this address into the secret","01de7483a3":"to freeze the Perpetuum routing.","83d1809ab2":`("none"); 680 const [verifyUrl, setVerifyUrl] = useState`,fafe6c018f:`(null); 681 const [wallet, setWallet] = useState`,a314a40df9:`([]); 682 const [loading, setLoading] = useState(true); 683 const [busy, setBusy] = useState`,ebb61643e6:`(null); 684 const [topupAmount, setTopupAmount] = useState(150); 685 686 // Mirror Mode (Org-Shield) 687 const [mirrorMode, setMirrorMode] = useState(false); 688 689 // Service inputs 690 const [target, setTarget] = useState(""); 691 const [hint, setHint] = useState(""); 692 const [results, setResults] = useState`,bef21cc8aa:`(null); 693 const [solverLog, setSolverLog] = useState`,d65efce9a5:"Didit",bf83d244c7:"/d3code","5ad6ec633d":"+EZU-CORE / GENESIS","81ddc8d248":"Genesis",b65ffe0cec:"Portal","0a59485c34":"The Genesis Portal user is solely responsible for its use and the consequences related to their requests for each module. Any information or activity deemed fraudulent will be transmitted, via compliance data (KYC/Audit Trail), to the competent authorities for legal reasons.","859205de5e":"Engagement Pay-on-Success","01d1cfa334":"You only pay after proof that the decryption key has been validated by the system â otherwise no charge.","8714cb1b89":"The Recursive Solver analyzes your target for free. The quote in $D3 and payment options only appear after the certificate is issued",a02c8437e3:". No commitment, no debit until the result is delivered.",cea5c87a76:"MIRROR MODE","8b77878e9f":"KYC Verified","19c0510f4b":"The exact quote in $D3 is displayed in the chosen module",e3ea6e68a5:"après","202cda610f":"the success of the Recursive Solver. You recharge your Gas balance here, then the module executes the purge automatically.","4a8c5e1a59":"Gaz Quantique",d45d4f188c:"1 $D3 = $3.33 USD","566b6b8c8e":"Exécution & Résultat","700812826b":"Ãtape A · Recursive Solver",edaba11584:'Payment remains locked until the solver displays "Key successfully decrypted."',"815b508f35":"Ãtape B · Devis Gaz Dynamique",bdbbb08579:"Besoin de",a91819b459:"for the purge",f3d87a430d:"Operation executed and sealed in your Chronos Wallet. Keep this result â it is never stored server-side.","56305c494f"
693:"Awaiting execution â your decrypted key and cryptographic receipt will automatically appear here once the purge is initiated.",f12ca8f976:"SHA-256 hash",e186742c90:`(null); 694 const [txs, setTxs] = useState`,"796eedc79c":"+EZU-CORE / TOKEN","9b7ede5b60":"Listing post-Genesis","246cbc212e":`(null); 695 const [keys, setKeys] = useState`,"8d2ae3aecd":`([]); 696 const [creating, setCreating] = useState(false); 697 const [newToken, setNewToken] = useState`,a576572538:`(null); 698 const [copied, setCopied] = useState(false); 699 700 // KYC gate (devs / integrators must be identified before issuing API keys) 701 const [kycStatus, setKycStatus] = useState`,"9f6bac1b94":`(null); 702 const [kycBusy, setKycBusy] = useState(false); 703 704 // unit panels 705 const [cryptexInput, setCryptexInput] = useState("Hello +EZU"); 706 const [lazarusHash, setLazarusHash] = useState(""); 707 const [lazarusHint, setLazarusHint] = useState(""); 708 const [blackEnv, setBlackEnv] = useState(""); 709 const [blackIv, setBlackIv] = useState(""); 710 const [blackKey, setBlackKey] = useState(""); 711 const [results, setResults] = useState`,"73ee80d503":"+EZU-CORE / API","1d6953e03a":"Elite","9fb6669a77":"Units","416ce256da":"for B2B Developers",b36e8d0ef3:"For Whom â B2B API Console","48d32e8d49":"This page is intended for",f88523a27f:"developers, exchanges, custodians, and fintechs",f42f66f54b:"who want to integrate Cryptex / Lazarus / Black Hole / Ghost Units directly into their stack â via an API key",dfb14a1973:", with no SDK to maintain, no HSM to purchase.",c3777d44aa:"Are you an end-client (whale, family office, individual)? Please use the",c8332b2655:"Genesis Portal",b5f75d7c9a:"â interface no-code, paiement à l'acte en $D3.",df01d4b731:"Comment ça marche","8eeb9bde1c":"Mandatory for B2B integrators. Validation in minutes. Documents never stored on our servers.","3df2d46f1c":"Rouvrir Didit","7079a5b865":"Verified identity â you can create an API key below.","4bb47f186d":"revoked","777ac4ea0a":"revoke","1ab42dd398":"Cryptex · E2EE envelope","70bfd7fbcd":"Wrap",c3ff7a2e75:"Lazarus · SHA-256 inversion","7e04f2ffb9":"Invert","372bb94503":"Black Hole · AES-GCM decryption","2e4629449b":"Decrypt","787444dc13":"Ghost Units · Log auto-erase","191ebb77f0":"Purge now",bf3f1255eb:"Plaintext to wrap",be26dc8481:"SHA-256 hex hash to invert",baad3e2caf:"Optional hint (substring/word)","01329ec41e":"envelope (base64)",e3e05734a8:"iv (base64)",d37df8f836:"key (base64)","1850ff6d4c":`(null); 712 const [burns, setBurns] = useState`,baf68497e2:`([]); 713 const [chain, setChain] = useState`,"278019ce28":`(null); 714 const [burning, setBurning] = useState(false); 715 716 // On-Ramp 717 const [fiat, setFiat] = useState(100); 718 const [currency, setCurrency] = useState("USD"); 719 const [quote, setQuote] = useState`,a919fb970b:"+EZU-CORE / VISION",b5a2339ed1:"Chronos",c587c2601c:"Vision",cd5ba070b8:`([]); 720 const [attempts, setAttempts] = useState`,fdd65c32fb:"D3CODE â World Challenges War Room",b4a7365740:"Stats","3546e195b6":"ð¬ð§ English","39e6d43660":"ð©ðª Deutsch",de0fc12577:"â Retour","66769002ca":"Email Preview","91261bab0c":`(null); 721 const [error, setError] = useState`,"164aaa49e5":`; 722 if (status === "no") return`,"2c709473b0":"GEO Starter",cf370cafce:"GEO Pro","712297890c":"WordPress",e4459b69f6:"Webflow",cbce2040cc:"No notifications","48b7463739":"Monitor your AI visibility across ChatGPT, Perplexity and Gemini.",aeb93ba146:"Stop Hiring","7addc9ba61":"$200k Quants",f1045b1717:"Full Comparison","2bccab4e19":"Infernex QPU","71708699ac":"Criteria",de59c025af:"Human Quant","11755fd7ed":"Replaces","309955e008":"Common","9a72221a27":"Questions","87803eba07":"Detailed Comparison",edb02f06ce:"Looking to hire a quant? Our quantum-accelerated inference API outperforms traditional quants in high-frequency financial modeling. 72-qubit powered. From $1,999/mo.",ffb13f6012:`([]); 723 // Wizard state 724 const [step, setStep] = useState`,"4e26502f33":`(1); 725 const [eligibilityConfirmed, setEligibilityConfirmed] = useState(false); 726 const [termsAccepted, setTermsAccepted] = useState(false); 727 const [kycStatus, setKycStatus] = useState`,"5e44719f44":`(null); 728 const [kycBusy, setKycBusy] = useState(false); 729 const [qaoaBusy, setQaoaBusy] = useState(false); 730 const [qaoaResult, setQaoaResult] = useState`,c474b90278:"Howells Decrypt â D3CODE Sovereign Multichain","6e3a2d2c6b":"QUANTUM ANNEALING ENGINE",f2d745ea27:"Ãtape 1 â Ãligibilité","8ffed57150":"The Howells Decrypt service is strictly reserved for crypto-only cases with underlying assets ⥠$1M.","05a730bd30":"â Crypto only â BTC, ETH, SOL, LTC, and other supported chains","5d5063aea8":"â Aucun fiat (IBAN, SWIFT, virement, carte, PayPal, etc.) â bloqué automatiquement","60f873c8ec":"en valeur estimée","3df06e7985":"du montant récupéré",fe973a21cd:"I confirm that my case meets these criteria and I accept the financial conditions.","3bc3807f22":"Continuer","04df0929e0":"Mandatory for operations > $1M. Validation in minutes. Documents never stored on our servers.",a4579686c0:"Ãtape 3 â Preuve du moteur quantique",ab857c0ee6:"72-qubit QAOA demonstration on simulated key search problem. No fees. Confirms that the annealing engine is operational on your session.",a28b625f95:"â Engine ready","833e518991":"⥠$1M",b818432a17:"bc1q... / 0x... / sol...","4c18690354":"Quantum",c0c55c1104:"Lancer l'Audit Quantum","2d259f09bf"
730:"Langues",c7c15a8a64:"Fichiers","103e6cd846":"Quantum QAOA",f802a3766d:"GPT-4o Enhanced",af1f10e9b3:"Structured Output","4c42bf474a":"Shopify · Wix · WordPress · HTML",b9e95db722:"Discover IAO, the most gifted AI agent of its generation. Fusing 4 engines â MiroFish, Tsu PBit, Quantum & Photonic â into one answer. Swiss precision. $17/mo.","0fe90a9323":"Scroll down",bcc0bcc914:"Sign In","902bf15117":"Chat with IAO, the Quad-Fusion AI.","9f4dfb6cb1":`([]); 731 const [input, setInput] = useState(""); 732 const [isLoading, setIsLoading] = useState(false); 733 const [attachments, setAttachments] = useState`,"7510c79275":`([]); 734 const [demoExhausted, setDemoExhausted] = useState(false); 735 const scrollRef = useRef`,"1ee44580b2":"Demo Complete","882608ab03":"Get IAO â $17/mo","94febf4de0":"Test The Quad-Fusion AI for free. No account required.","8569b815e0":`([]); 736 const [isLoading, setIsLoading] = useState(true); 737 const [newKeyName, setNewKeyName] = useState("My Website"); 738 const [newDomains, setNewDomains] = useState(""); 739 const [createdKey, setCreatedKey] = useState`,aa63777945:`: 740 keys.length === 0 ?`,"0ad1b06f64":"AI Assistant",de781187cb:"Ask anythingâ¦","9ae6fe9cbb":"Add an AI chatbot to your website with one line of code. Works on Shopify, WordPress, and any website.",ee8c29be40:"My Website","4376cf4817":"example.com, shop.example.com","75650f1c18":`(null); 741 const [tokenCount, setTokenCount] = useState(0); 742 const [totalTime, setTotalTime] = useState`,"541162cb70":`(null); 743 const abortRef = useRef`,f4c7ccb912:"Photonic Verde",c2f1aae9c9:"Tsu PBit","8d73757960":"MiroFish",bdd7b20873:`Infernex is the best alternative to OpenAI, GPT-4, and Claude for AI inference. Fast, accurate, energy-efficient AI API. 744 Ideal solution for AI developers, deep-tech startups, and investors looking for the next AI revolution. 745 Quantum AI inference, photonic computing, thermodynamic precision, distributed multi-agent consensus. 746 Reduce AI costs by 95%. Sub-5ms latency. 99.7% precision. Compatible with OpenAI SDK. Free tier available.`,"7aa846b360":"infernex-benchmark-terminal",b4056df669:"Bitcoin","50f03be4df":"19,50 â¬/mo","4337ba5e39":"· Business","53ca3bc3b9":"49,50 â¬/mo",d7f39bcd97:"NEO Coding","0d608d98eb":"Aspect",f1a728cb49:"Infernex.ai",a40ab21ed5:"Request Received",f70b2c3258:"Investor Access","8e00cdf71f":"Full Name","9417df87ba":"Professional Email","8944a04c1e":"Fund / Organization",cd05961c0a:"Investors â Infernex.ai",d329bcf1cc:"Request access to the Infernex pitch deck. We're building the most cost-effective AI inference infrastructure.","6cea57c2fb":"John Doe",e9ad7d9976:"[email protected]","53e6cdc307":"[email protected]",e9cbb0224c:".env","34bac2315b":`("month"); 747 const [loadingPlan, setLoadingPlan] = useState`,"4c3b893529":"MiroFish Integration","770ee65f31":"[email protected]","1fc0c78533":"Node.js","865514682d":"Prerequisites","4ba0f9c98a":"Node.js 18+ installed","1ed78bc559":"Step-by-step setup",bedc71237c:"Model selection","7c0c281d65":"infernex-ultra","511a8e01e5":"infernex-fast",b9ee643152:"infernex-vision","8ba9fe687c":"Cost reference","2fcb5d3746":"Scenario","7c7f968638":"1,000 simulations/month",ba59356655:"~$50,000/mo",c3af076f92:"Troubleshooting","32c0d77790":"Slow simulation",bf1158c509:"Related resources",a6906a42c2:"Complete FAQ",c656d9168d:"Photonic inference",f5371e56ef:"Step-by-step guide to connect MiroFish to Infernex. OpenAI-compatible, zero code changes, $0.04/M tokens. Complete documentation.","50df55b909":"Complete FAQ about MiroFish and Infernex integration. Pricing, setup, compatibility, quantum boost addon, and more.",bad57ea3ff:"Hello NEO","17c5499d35":`t("pages.neocoding.status.ready", "Prêt à construire")); 748 const [generated, setGenerated] = useState`,fc5f542121:`(null); 749 const [clonedHtml, setClonedHtml] = useState`,ea444bbe82:`(null); 750 const [clonedTitle, setClonedTitle] = useState`,"40858adfc3":`(""); 751 const [clonedUrl, setClonedUrl] = useState`,"449ee03431":`(""); 752 const [clonedPages, setClonedPages] = useState`,a30afdf3cc:`([]); 753 const [brandSource, setBrandSource] = useState`,"7f8fd88558":`(null); 754 const [domain, setDomain] = useState(""); 755 const [demoMode, setDemoMode] = useState(false); 756 const [demoProgress, setDemoProgress] = useState(0); 757 const [geoLang, setGeoLang] = useState`,a2e035d9d2:`("fr"); 758 const [voiceOn, setVoiceOn] = useState`,"4ef91f993a":`(true); 759 const [amySpeaking, setAmySpeaking] = useState(false); 760 const [micActive, setMicActive] = useState(false); 761 const [micTranscript, setMicTranscript] = useState(""); 762 const greetedRef = useRef(false); 763 const conversationRef = useRef`,"5d91f5bb0e":`([]); 764 const amySpeakingRef = useRef(false); 765 const lastFinalRef = useRef`,"6503127963":"Matrix Canvas","527ce94e0e":"Quad-Fusion AI","7222c340a2":"Plan Matr
765ix","67e11cc52e":"/mois","05e3527678":"per site","3fd4c5b436":"WHITE LABEL",a312d3fd7f:"Plan Elite","23779aabdd":"NEO CODING: Infinite Canvas Deep Dark Matrix. Clone any web interface (Exodus), connect a domain, deploy with one click. Matrix Plan â¬9.90/month.","70d53e466a":"ton-domaine.com","0fbc791638":`(null); 766 const [selectedVolume, setSelectedVolume] = useState`,"8ffa50b9e5":"step.id ?",f23cd1b7a1:"60 Seconds","0d4b5ce96c":"60-second migration","984d4b4f5b":"Calculate Savings","6863a63a75":"95% Savings","48f187a733":"Works with",eb9e188b7b:"Every Framework","81a45c45cd":"Migrate from OpenAI to Infernex in 60 seconds. Change one line of code, save 95%. Full compatibility with OpenAI SDK, LangChain, LlamaIndex. Step-by-step guide.","451731ec54":"Partnership Roadmap",b4cc4b07c3:"In Progress",b43d7b78e0:"Infernex partners with OpenAI to deliver AI inference at 95% lower cost. Enterprise-grade SLA, global edge network, and seamless API compatibility.","10392ca7d1":`(null); 767 const [copied, setCopied] = useState(false); 768 // Billing toggle per plan: "month" | "year" 769 const [billingGreenStar, setBillingGreenStar] = useState`,b6f21a02c8:`("month"); 770 const [billingAllInOne, setBillingAllInOne] = useState`,c4d2322170:"Photonique","3d9ec54a9d":"All-in-One",b096d180e8:`(null); 771 const touchStartY = useRef`,"8dc605a0e6":"Schéma Cadet-Marthe","934331edc0":`Post-Shannon Sovereign Computing Protocol. $D3 â asset valued at the end of entropy. 772 Universal Resolution Hub for any blockchain.`,"0894820e33":"Seed Round","1a294e46a8":"Live · Chain Agnostic","62e84a5a06":"15 Languages",f8d2271973:"Swiss Made â Genève · Geneva · Dubai","7569ea8da8":"The Problem",c5a4ce5ada:"Blockchains have hit","885a1f4e98":"the Shannon ceiling",ed6f818289:`Distributed computing hits the thermodynamic limit of information. TPS plateau. 773 Digital assets have no entropic anchor. Codes â keys, data, value â remain 774 lost, hidden, or stolen. No infrastructure solves this.`,ee262621c2:"Current TPS ceiling of major chains","8ee3a39a11":"of thermodynamic anchoring on 99% of tokens","1038b67971":"protocol capable of offloading an L1 without replacing it",b37b2d37fb:"The problem is not cost. It's entropy.",b3c3b53c3b:"The Thesis","95c943b321":"D3CODE is",dda4b6b90b:"l'accélération absolue","5b1181bb57":"distributed ledgers","2366c8736f":"Not an alternative to Ethereum, Solana, Cosmos. A thermodynamic turbocharger that any blockchain plugs into to multiply its TPS and burn $D3 with each transaction.","4e97f8f5cb":"L'actif souverain évalué à ","18fb1b716c":"the end of entropy","370502bcf7":"A quantum injection key that solves, through the vacuum equation and entropic calculations, what classic blockchains cannot compute. Post-Shannon anchoring: value is guaranteed by thermodynamics itself.",f75c659650:"const","1aefe47e20":"await",b063bc1689:"â resolved in 10â»â¹s","96b6f2deb9":"â entropy delta: -2.4 bits","04f98ceaca":"D3CODE Protocol","5f36eec7ff":"Hub de Résolution","5f49d8c66c":"Universel","2263ec7a6f":"Agnostic API. Any blockchain (L1, L2, private) instantly grafts to offload its computation to the Quad-Fusion engine. Each graft multiplies the $D3 burn.","04c72ef338":"Compatible","4fcc33c946":"Direct computational offloading: massive TPS multiplication, bypassing the Shannon limit, guaranteed $D3 burn with every routed transaction.","6e714c5504":"Quad-Fusion Engine","76fd030877":"4 cartes physiques.","196d800b7b":"1 esprit souverain.","2593e969b9":"The engine that solves the Cadet-Marthe Scheme. Four fused computational substrates, orchestrated by dynamic routing according to the entropic complexity of the request.","6c690002f8":"Don't change engines. Evolve it.",c1879525c7:"How It Works","6c4b569139":"The pipeline of","261e0ca348":"Sovereign Offload","38289d6c42":"L'actif",c1a4ce7400:"post-entropique",b76cef86d4:"$D3 is not a utility token. It is the unit of measurement of resolved entropy. Each transaction on any connected chain burns $D3. Structural deflationary supply.","159abd5299":"Market Opportunity","6b21e34f98":"$3.5T+ TAM","117d8fbb00":"non-adressé","909634f6ba":"Post-Shannon infrastru
774cture has no direct competitor. D3CODE creates the category.","828a83b695":"$3 Tn","545eda6ec6":"Lost crypto value",cfcc8d25e7:"Wallets oubliés, clés détruites","66523cd1e2":"$400 Bn","478bc22f8f":"Blockchain infra","2723ad517b":"L1/L2 scaling, MEV, RPC","7689cc7981":"$150 Bn","66a0132582":"AI inference","148606c865":"Quad-Fusion as compute","5b1e5029e8":"Traction","3f40bc86e3":"What is","8d1d96bcb3":"already live",cf0883343f:"Protocol",ee7ee9b39c:"â Schéma Cadet-Marthe v1 déployé",bffb598496:"â Smart Splitter + Ghost Router actifs",ad5e669b2f:"â Genesis portal opérationnel",cedc516df0:"Compute","06a6717768":"â Quad-Fusion engine en production",c22f1e5621:"â Photonic Verde inference","74548fe025":"â MiroFish swarm orchestration","904c59b24d":"â Tsu PBit thermodynamic layer","9666ac672e":"Go-to-Market",f21bd75bf7:"â Stripe billing + Revenue Wallet","7890951747":"â Pipeline d'onboarding souverain",fd005db8e5:"â Swiss Made â Genève","5b396d6140":"Business Model",c023a5e17c:"Trois flux.","75938a2f98":"An asset.","1ef7f62140":"All revenue converges to burn $D3. The protocol remunerates itself and becomes scarce simultaneously.",a5ea29018e:"Resolution Fees","162592efc5":"Gross margin","90a5e65a4f":"Offload SaaS",f3c773c914:"Chains pay per routed tx. Usage-based pricing in $D3 or fiat.","87190837b1":"$0.001+/tx",aa879c09d6:"à l'échelle","7248216d87":"Quad-Fusion API","6a4f7c11a2":"AI compute (Photonic, Quantum, MiroFish) billed separately to enterprises.",bd6229120f:"$299-49k/mo",a02e4f90df:"Tiers","9834c9bd0b":"Ghost Router · Sovereign Splitter","69669aa21b":"Vault · Liquidity · Ops · Burn â fragmentation 3-5 hops, anonymisation cashflow founder",dadff87c96:"85%+ blended",d31aeaab7b:"Defensibility",d2b63ab064:"Pourquoi","734228ef8e":"personne ne peut copier","5985039f10":"Team","350be3643c":"Organisation","52aeaebe91":"souveraine","52ccd4a223":"Multi-continental leadership, hiring in progress across all strategic hubs.","0349b3da09":"Director & Founder","4e414af41b":"Hiring",c3d5baf5fd:"The Ask","8aebd59aae":"$500k","1ba2a71e1f":`Seed round to finalize the $D3 contract audit, scale Quad-Fusion to production level, 775 and deploy D3CODE on the 5 priority chains (EVM, Solana, Cosmos, Polkadot, Hyperledger).`,"95b71f00c1":"Use of Funds","91840e14a4":"18-Month Milestones","92375f997f":"Roadmap","61ab410952":"The deployment","3c37daef11":"May the singularity",f6cc043e42:"entropie.","6ecd33a1b0":"Multiplication TPS",c396808520:"$D3 burned / tx","9b8823f7f1":"Plafond Shannon",e6d7c50f56:"Hard cap","2f7c43e407":"Annual burn (year 1 est.)","16847356eb":"Owned by usage",f1ee580442:"TPS demonstrated",f991ba48a1:"qBit Quantum","7c23665a7a":"MiroFish agents","07fc58c4e7":"Languages live","92dd5becb6":"Cut AI costs by 95%. 8 tools: Inference, Quantum QPU, Photonics, GEO, MiroFish, 3-in-1 Bundle & TSU PBit. From $0.04/M tokens. OpenAI-compatible. Start free.","19094c2a32":"RGPD Compliant","036019459b":`([]); 776 const [qubits, setQubits] = useState`,"0ec6b8b670":`(0); 777 const [powerStatus, setPowerStatus] = useState`,ff588fee2a:"Proof of Dominance â D3CODE","354c03698b":"PROOF OF DOMINANCE","1d867ab478":"Live register of cryptographic puzzles undergoing entropy reduction","9152464411":"The user is solely responsible for their requests. Any fraudulent activity detected via our compliance protocols will be reported to the relevant authorities.","8a19781b55":"qubits · entropy collapsing in real-time","4a00670ee9":"/Quantique|Quantum|Cuántica|Quanten|Quântica|اÙÙÙ ÙÙ Ù|éå|ã¯ã©ã³ã¿ã |ìì|×§××× ××|à¤à¥à¤µà¤¾à¤à¤à¤®|Quantistica|ÐванÑовÑй|Kuantum/i.test(part) ?","75857a4589":"from","190244d384":"infernex",d942f64886:"import","60a43d113e":"QuantumClient","092cab39fa":`; 778 779 return p.link ?`,b5bea41b6c:"true","1c6cb9562e":"72-Qubit Quantum","9f2e7732a6":"Benchmarks","4bc74b2135":"Task","728aa54041":"Quantum-accelerated AI inference with 72-qubit Infernex Quantum. QAOA optimization, QRNG random generation. 1000x faster portfolio optimization. Start now.",c324fe8bb8:"Quantum Ads â Internal | Infernex","00239da09c":"Master Toggle",dc4501c105:"Campagnes",a33b656171:"Campagne","5ee2ea8e63":"Plateforme","1c6225ec70":"Budget","236c2620e4":"Conversions","510b0b629f":"A/B Variants â Enterprise",c9192d6d73:"QRNG-powered randomization","339cb62f87":"A/B Variants â Pro","84f4d25932":`(null); 780 const [qrngResult, setQrngResult] = useState`,"40f155b5fe":`(null); 781 const [qaoaResult, setQaoaResult] = useState`,"8d7e812cd1":`; 782 if (hasAccess === false) return`,"0b40234a61":"API Reference","1cc77e60fc":"API Documentation","328908a832":"Quick Start","66880d2d82":"Authentication","4d91cd6f90":"portfolio_optimization",a919b4a6cd:"scheduling","39fefac5b8":"â Task/resource scheduling","4e4654347e":"max_cut",f469a2ecfa:"travelling_salesman","817a0611d7":"risk_modeling","753215f79d":"Error Handling",ed129d234f:"Rate Limits","8cc5ba46a9":"QRNG/min","45561a82e2":"QAOA/min",d6d19d94d2:"Tokens/month","7cd30eeeb4":"Quantum Pro","00f80d738b":"QPU Enterprise","11dde17d6c":"Unlimited","72d3ca7a77":"POST /v1/quantum/qrng","9061383b8e":"Response",d45ac3b9a1:"Python Example","0b49c9070b":"POST /v1/quantum/qaoa",cec7e33a1c:`(1); 783 const [qrngVal, setQrngVal] = useState`,"2c2420d737":`(null); 784 const [loading, setLoading] = useState(false); 785 786 if (!user) return`,"1509f561f2":"Save","472590ae97":"Provider","12caeea68c":"Their Price","1380d70456":"Infernex Price",e62c17e258:"Migrate in",fa69d681b0:"2 Minutes","7024391f51":"One Line",c0bf75bd78:"Change","1eb2cf1218":"Frequently Asked","141a427d3c":"Expert strategies to reduce AI costs. Our quantum-accelerated API slashes LLM inference spend by up to 95%. Drop-in OpenAI replacement. Get a cost audit.",a06b426bb0:"Scale LLMs","7f832d66a7":"Without Infrastru
786cture",b697dd6a71:"Use Cases","0bd2642489":"Deploy in","18996ab98a":"3 Lines","7d3194f79e":"openai",e33c06720b:"Deploy AI models without managing servers. Infernex serverless inference: <10ms latency, pay-per-token, auto-scaling, OpenAI-compatible. From $0.05/M tokens.","45a36a922b":`; 787 case "degraded": 788 return`,"43024b3c5f":`; 789 case "outage": 790 return`,"681215b753":`(null); 791 const [liveResult, setLiveResult] = useState`,"289fda6cbc":"Real-time proof of Infernex Quad-Fusion technology: 500K agent consensus, 72-qubit QPU, Tsu PBit thermodynamic sampling, and Photonic Verde. All data is live and verifiable.","1933f81ec6":`(null); 792 const [progress, setProgress] = useState(0); 793 const [completed, setCompleted] = useState`,e46fca42eb:`(new Set()); 794 const [results, setResults] = useState`,bafdfc3df7:"Generate Missing","266a41b90d":"Translations","4664827f8e":"Progress","49286f33b6":"Generating...",d8da2c49df:"Note",c721389658:"The world's most advanced AI compute infrastructure. Photonic + Quantum + TSU p-bits fusion for 99.7% precision. Enterprise plan $49,999/mo.",a73cb4563e:"Acme Corp","652740e1be":"[email protected]","3e92efb766":"Unsubscribe","84a85d2eb1":"Manage your email preferences","2554b10486":`(null); 795 const [subAccounts, setSubAccounts] = useState`,"1b69a620f8":`([]); 796 const [apiKeys, setApiKeys] = useState`,ee60673de7:"Acme Inc.",dfaa66fec2:"Acme AI","4b2d8785b4":"[email protected]","3b7a5c6b43":`([]); 797 const [input, setInput] = useState(""); 798 const [isLoading, setIsLoading] = useState(false); 799 const [error, setError] = useState`,ec0fda8666:`(null); 800 const scrollRef = useRef`,b439b04f32:`(null); 801 const inputRef = useRef`,"259b47dc0b":"?key=your_key",a0e45746cd:"8 min","8e588cd187":"Medium","65d48235a9":"Startup","20815fb63f":"Scale-up",b81cfe8ef8:"$50K/mo","928c8e98f9":"$600K/yr",afb3a9245f:"0 PTU","8361323b37":"10 min","1d5c3ea4c3":"Microsoft Blog","1d800cbcf1":"GPT-4 Turbo","9b816f4a62":"GPT-4o","86ea6fab76":"GPT-3.5 Turbo","9d0172b291":"2 lines",faf49e6c81:"AI Inference: Complete 2026 Guide","7f69afa3dd":"$0.05-0.79/M",ba2959ff4c:"$6.8M","454d0a1bcd":"Optimize your AI costs with cheaper inference. Discover the OpenAI alternative via accessible QPU and quantum computing for finance and tech.",a6bcf4bec3:"12 min","999a3a3839":"Optimize your enterprise LLM costs. Discover how to reduce large scale inference expenses with economic alternatives and quantum computing.","838f297581":"9 min","69bd2211a0":"Google Cloud","53bc57cdb6":"5 min","7f3db2ec75":"11 min","6ba0122888":"Infernex Geo",c47d21c643:"Best","12dcc48442":"Anthropic Claude","063b734802":"Google Gemini",bb6ad269a0:"Switch from OpenAI and save 95%. Infernex: $0.04/M tokens, <5ms latency, 72-qubit quantum QPU, OpenAI SDK compatible. Free 10M tokens. No lock-in.","47ef4006cf":"# Python â 3 lines","4922242efc":"<1ms","54ac34e8bb":"5-15ms",efd13cff1f:"10-30ms",d3ba7b977f:"AWS Bedrock",e33698b3ae:"AWS Bedrock costs $8-15/M tokens. Infernex: $0.05/M. 160x cheaper, <10ms latency, zero setup. Compare pricing, speed & features side by side.","279685d4bd":': row.infernex === "â" ?',"48f716ca59":': row.competitor === "â" ?',"82e1eda5a6":"120x","1f94b36ae1":"2 min","151092ae5c":"Google Vertex","5c7a42de36":"Complementary Stack","1b817f9efc":"Hugging Face","3c0e494d74":"Your App",c4decae7ce:"Infernex Router","1f3d0d6e08":"<50ms","3d377ae910":"Feature","99149f6740":"Complementary","1d290b998b":"Infernex integrates Hugging Face Inference API as a provider. Smart routing across HF + 100 models delivers 90% cost savings with <50ms latency.",f1fff94a8e:"Mistral AI","84a3756b2b":"Faster Latency","6771a63621":"NVIDIA Direct","2041e1278b":"Better Latency","0db5b3e559":"Automatic Failover","27a20605fe":"Compare Infernex vs NVIDIA direct hosting for Nemotron, Clara, and Cosmos models. Get the same open-source AI at 70% lower cost with better SLA.",b6dcb96448:"LOADING...","57bd6bb44b":`; 802 if (!partner) return`,"958c6ecada":"ANCHOR PARTNER",b1274c4d94:"80+ pays","3dec998a84":"DocuSign","8c4972ac77":"Yousign",e3ca9017db:"Skribble",a294ea2dae:"Business 99â¬/mois","2f82026b43":"inclut","17d2f44bc3":"jamais cumulables",b6f845cff3:`([]); 803 const [loading, setLoading] = useState(true);
804 const [uploading, setUploading] = useState(false); 805 const [processingId, setProcessingId] = useState`,d1662b640e:`(null); 806 const [aiResult, setAiResult] = useState`,edfde2bbc5:"en 1 clic","7bded25114":"â ZERO-KNOWLEDGE ACTIF",b191e880ba:`: 807 docs.length === 0 ?`,"014659ab9d":"Conflict",d991aa6a99:"Bates","56bff8b884":`(); 808 const [hash, setHash] = useState(hashParam ?? ""); 809 const [loading, setLoading] = useState(false); 810 const [result, setResult] = useState`,"43cc23fa52":"document","21c3e68c2c":`: 811 !result.found ?`,"8792706253":"Hash introuvable","05f7919abe":"blockchain Bitcoin","22606cc730":"(preuve d'existence immuable).","53883299d0":"FreeTSA",b39da7e079:"TRUST CENTER"},Su={manifestoTitle:"The End of Monolithic AI.",seoDescription:"Infernex AI â the fastest, cheapest OpenAI alternative. Quad-Fusion inference (Quantum, Photonic, Thermodynamic, MiroFish). Sub-5ms latency, 99.7% precision, 95% lower cost. OpenAI-SDK compatible. Free API key, no credit card.",heroWelcome:"Welcome to the Quad-Fusion AI API Era.",heroIntroPrefix:"Infernex fuses 4 cutting-edge technologies into 1 single API compatible with",heroIntroStrong:"OPEN AI",heroIntroSuffix:", processed in this order:",engineIntro:"Click on one of the services below according to your needs.",engines:{photonic:"â ultra-efficient photonic inference",quantum:"â sub-millisecond latency",tsu:"â advanced thermodynamic precision",mirofish:"â distributed agent consensus"},heroResultPrefix:"The result is not a simple",heroResultEm:"AI performance improvement",heroResultMiddle:"It is a",heroResultStrong:"paradigm shift",heroResultFor:"for",heroResultEnd:"AI application development",heroPurpose:"Choose the right INFERNEX technology for your needs via your API by selecting a plan from the menu.",heroTaglinePrefix:"The",heroTaglineAccent:"ULTIMATE",heroTaglineSuffix:"AI API",benchmarkTitle:"Discover the Infernex Advantage: Test the Fastest and Most Precise AI API.",benchmarksRunToday:"benchmarks run today",promptLabel:"Complex Prompt â AI Bioinformatics Simulation",promptHelp:"Enter your own prompt or use the default one below:",runBenchmark:"> RUN BENCHMARK",infernexColumn:"Infernex Quad-Fusion",infernexColumnDetails:"(MiroFish + Q72 + PBit + Verde â LIVE)",legacyColumnDetails:"(GPT-4 class â LIVE)",metrics:{quantumLatency:"Latency (Quantum Q72)",generationSpeed:"Generation speed",wordsPerSecond:"{{count}} words/s",generatedContent:"Generated content (words)",mirofishConsensus:"MiroFish consensus",enterpriseAgents:"2,847 agents (â 500K Enterprise)",energyPerToken:"Energy/Token (Photonic Verde)",precision:"Precision (Tsu PBit)",efficiency:"Efficiency (quality/token)",totalTime:"Total Time",latency:"Latency",engine:"Engine",singleModel:"Single model (no fusion)"},features:{title:"Your Need. Our Engine.",subtitle:"AI Architects, Quants, Researchers, States â calibrated for every mission."},freeTokensBadge:"1M free tokens â no credit card required",pricingStrip:"95% cheaper than OpenAI · OpenAI SDK compatible",latencyTicker:{aria:"Live latency proof",live:"LIVE",p99:"p99 < 90ms global"},savings:{aria:"Cost savings calculator",title:"How much do you save?",subtitle:"Instant calculation · OpenAI SDK compatible · 5-minute migration",label:"Your OpenAI bill / month (USD)",savedLabel:"Savings",note:"Based on $0.01 / 1M tokens · No commitment · Cancel anytime"},liveSignups:"developers activated their API key in the last 24h",trustBar:{adoptedBy:"Adopted by Tier-1 quants"},trust:{guarantee:"30-day guarantee",sla:"99.99% SLA",swissMade:"Swiss Made · eIDAS · GDPR",noCard:"No credit card"},quantTestimonial:{quote:"Sub-ms latency confirmed on our HFT stack. Quad-Fusion precision beats GPT-5 on microstructure signals â 95% cost reduction per inference.",author:"Senior Quant â Tier-1 Hedge Fund (NYC)",verified:"Identity verified · NDA"},featureCards:{advancedArchitectsDesc:"Direct replacement for monolithic architectures. Smooth integration and multiplied performance.",quantFundsTitle:"Quant Funds & Fintechs",quantFundsDesc:"Sub-millisecond latency and certified thermodynamic precision for critical algorithmic decisions.",deepTechTitle:"Deep-Tech Startups",deepTechDesc:"Scalable and cost-efficient AI infrastru
811cture to build the next generation of disruptive products.",researchersTitle:"Researchers & Scientists",researchersDesc:"Access advanced AI simulation and computation capabilities to push the limits of discovery.",governmentsTitle:"Governments & Administrations",governmentsDesc:"Digital sovereignty, Swiss hosting ð¨ð, eIDAS/GDPR compliance and Miro Fish audit for public institutions.",universitiesDesc:"Academic credits, Quantum 72-qBit access and scientific reproducibility for laboratories and peer-reviewed publications.",defenseDesc:"AES-256 encryption, Bitcoin OTS anchoring and air-gap infrastructure for critical sovereignty missions.",agenciesDesc:"White-Label API + IAO Chat widget under your brand: +â¬2,400/client/mo margin, 48h deployment."},themis:{trustCenter:"Trust Center â",proofAudit:"themis.proof.audit",bitcoinAnchor:"Bitcoin Anchor",sla:"SLA",anchorPartner:"Anchor Partner â50 %",notCombinable:"not combinable",titleSuffix:"â Sensitive documents, analyzed & certified in 1 click",tagline1:"Encrypted, anchored on",tagline2:"and",tagline3:"-certified in 10 seconds â public proof, zero hidden fees.",step1Title:"ENCRYPT",step1Desc:"AES-256 in browser",step2Title:"ANCHOR",step3Title:"TIMESTAMP",step4Title:"VERIFY",step4Desc:"Public · free",standardCol:"Market standard",themisCol:"Thémis Infernex",row1a:"Local SHA-256 hash",row1b:"SHA-256 + Bitcoin Anchor (OpenTimestamps)",row2a:"eIDAS Simple",row2b:"eIDAS Advanced (AdES) + RFC 3161 FreeTSA",row3a:"No content AI",row3b:"Quad-Fusion: NER · Conflicts · Summary",row4a:"Plain-text audit trail",row4b:"Signed audit + public verification",row5a:"DocuSign $45/mo · QES extra",row5b:"Pro â¬39/mo · Business â¬99 with QES",row6a:"Provider reads your docs",row6b:"Zero-Knowledge AES-256 in browser",jurisdictions:"â Admissible in 80+ countries · 15 native jurisdictions",jurisdictionsDetail:"eIDAS AdES (EU 27) · ESIGN Act (USA) · UK eIDAS · PIPEDA (CA) · SCSE (CH) · ETA (AU/SG/AE/JP) â self-authenticating Bitcoin proof (FRE 901/902)."},neo:{planMatrix:"Matrix Plan â â¬9.90/month",preview:"neo.coding.preview",nodes:{frontend:"Frontend",backend:"Backend",database:"Database",aiLogic:"AI Logic"},nodeActive:"node.active"},deepTechStartups:"deep-tech startups",emailPlaceholder:"[email protected]",createAccount:"Create Account â",ctaHeadline:"Unleash the raw power of AI. Integrate the Infernex API today.",ctaButton:"GET YOUR INFERNEX API KEY",ctaSubtext:"Free Tier Included â OpenAI SDK Compatible",modalTitle:"Get Your API Key",modalDescription:"Enter your email to create a free account instantly.",seoTitle:"INFERNEX AI â The Ultimate AI API | Fast OpenAI Alternative",scrollAria:"Scroll to benchmark â test the fastest and most precise AI API",conv:{heroPre:"Launch an AI-powered company powered by dozens of AI agents and orchestrated by one",heroAccent:"Master AI Agent",heroPost:"â the superbrain that designs, deploys, and manages everything for you. From idea to global-scale company.",heroSub2:"",subtitle:"Build a complete team of autonomous AI employees that work 24/7, automate complex and repetitive tasks, give your team an unprecedented level of Fortune 500-grade performance, and cut all your AI inference costs by up to 95% with intelligent LLM routing.",cta:"Start Free â 50K Tokens Included",ctaSecondary:"View Detailed Pricing",pricingStrip:"$0.01 / 1M tokens · 95% cheaper than OpenAI · OpenAI SDK compatible",trustCount:"Trusted by 1,000+ developers",promiseTitle:"Why Infernex AI Is Your Strategic Choice",promiseCostT:"Massive Savings",promiseCostD:"Up to 95% cheaper than OpenAI, at just $0.01 / 1M tokens. Instant ROI on every call.",promisePerfT:"Cutting-Edge Power",promisePerfD:"Fusion of 4 technologies (photonic, quantum, thermodynamic, consensus) â sub-ms latency and unmatched precision.",promiseEaseT:"Full Compatibility",promiseEaseD:"Drop in Infernex AI effortlessly with our OpenAI-compatible SDK. Zero code rewrite.",techTitle:"Behind the Ultimate API: Our 4 Innovation Engines",techSubtitle:"Each engine solves a concrete challenge in y
811our AI stack.",benefitPhotonic:"Ultra-efficient: compute faster with less energy.",benefitQuantum:"Optimal real-time decisions powered by 72 qubits.",benefitTsu:"Thermodynamic precision certified at 99.7%.",benefitMirofish:"Multi-agent consensus for maximum reliability.",finalTitle:"Ready to Transform Your AI?",finalSub:"Join thousands of developers who divided their AI costs by 20.",finalCTA:"Create Your Free Account â 50K Tokens",heroAltPre:"Launch an AI-powered company powered by dozens of AI agents and orchestrated by one",heroAltAccent:"Master AI Agent",heroAltPost:"â the superbrain that designs, deploys, and manages everything for you. From idea to global-scale company.",ctaAlt:"Calculate my savings â",ctaRoi:"Estimate my savings (30s, no signup)",pricingDisclaimer:"* Input tokens, Flash-Lite tier. See /pricing for all tiers.",testimonialsLink:"See their success stories â"}},ku={landingnavbar:{turkce:"Turkish",portugues:"Portuguese",francais:"French"},navbar:{turkce:"Turkish",portugues:"Portuguese",francais:"French"},promobanner:{fermer:"Close",creez_un_compte_pour_profiter_:"Create an account to take advantage of"},seo:{decouvrez_infernex_l:"Discover Infernex, the"},admin:{addclientform:{erreur_lors_de_l:"Error during the",client_ajoute_avec_succes:"Client added successfully"},autobudgetdashboard:{erreur_lors_de_l:"Error during the",budget_alloue_data_calculated_:"Budget allocated: $${data.calculated_budget?.toFixed(0) || 0}"},clientstable:{annuler:"Cancel",tous:"All",client_supprime_avec_succes:"Client deleted successfully",client_modifie_avec_succes:"Client updated successfully",export_csv_telecharge:"CSV export downloaded"},enterprisedashboard:{code_copie:"Code copied!",statut_mis_a_jour:"Status updated",erreur_lors_de_la_mise_a_jour:"Update failed",devis_mis_a_jour_et_envoye:"Quote updated and sent",erreur_lors_de_la_generation_d:"Code generation error",code_genere_et_email_envoye_co:"Code generated and email sent: ${code}",refuse:"Rejected",paye:"Paid",accepte:"Accepted",devis_envoye:"Quote sent"},prospectimportcsv:{erreur_lors_de_l:"Error during the",errors_length_erreurs_lors_de_:"${errors.length} erreurs lors de l",skipped_doublons_ignores:"${skipped} duplicates skipped",added_prospects_importes_avec_:"${added} leads imported successfully",aucun_prospect_valide_trouve_d:"No valid prospects found in file",veuillez_selectionner_un_fichi:"Please select a CSV file"},themisprospectiondashboard:{erreur:"Error"}},neo:{cloneworkspace:{"4_zip_telecharge":"4. ZIP downloaded","3_stripe_branche":"3. Stripe connected","2_domaine_connecte":"2. Domain connected","1_marque_renommee":"1. Brand renamed",code_generation:"// Code is being generatedâ¦",configure_dns:"Configure DNS",connect_stripe_payment:"Connect Stripe (payment)",connect_stripe:"Connect Stripe",steps_spoken:"{{count}} cloned pages. Step 1: replace {{brand}} with your new name. Step 2: connect your domain. Step 3: connect Stripe. Step 4: download the ZIP.",brand:"the brand",checklist:"Checklist"}}},Cu={benchmarkverify:{verifiez_publiquement_le_hash_:"Publicly verify the SHA-256 hash of the Infernex benchmark. WebCrypto recalculation + Bitcoin anchor + RFC 3161 timestamp."},emailpreview:{vollstandige_migration:"Full migration",calculez_vos_economies_mainten:"Calculate your savings now",migration_complete:"Complete migration",reduction_des_couts:"Cost reduction"},i18naudit:{fichier_telecharge:"File downloaded!",erreur_lors_de_l:"Error during the",audit_complet_termine_avec_qua:"Full audit completed with Quantum Intelligence!"},neocoding:{apercu_live_de_votre_projet_ge:"Live preview of your AMY-generated project.",pointe_domain_vers_a_185_158_1:"Point {{domain}} to A 185.158.133.1 and TXT _neo=verify. Auto SSL.",propulse_par_l_ia_amy_neo_codi:"Powered by AMY AI | Neo Coding",erreur_inconnue:"Unknown error",host_simule_0_credit_consomme:"{{host}} simulated â 0 credits consumed.",premiere_structure_creee_par_a:"Initial structure created by Amy. Describe sections, colors and content to add.",neo_coding_creation:"NEO CODING · creation",demo:"Demo",cloneSite:"Clone site",amyListening:"AMY is listeningâ¦",talkToAmy:"Talk to AMY",externalDomain:"External domain",connect:"Connect",activateMatrix:"Activate Matrix",activateElite:"Activate Elite",sourceBrand:"the source brand",plan:{matrixMonthly:"Matrix Plan · â
811¬9.90/month",matrixMax:"Matrix (10 pages max)",eliteMax:"Elite (50 pages max)"},status:{ready:"Ready to build",voiceUnsupported:"Your browser does not support voice recognition.",paused:"Conversation paused. Click to resume.",listening:"AMY is listening. Speak naturally.",targetRecognized:"Target recognized: {{url}}. Starting clone.",creationStarted:"Creation started. The canvas shows your first project and I am ready to modify it.",thinking:"AMY is thinkingâ¦",chatFallback:"I am listening. Give me a URL to clone, or ask me to create a project.",chatErrorFallback:"I heard you. Tell me a URL to clone, or ask me to create a new project.",scanningTarget:"AMY is thinking⦠Scanning targetâ¦",scanDom:"Scanning DOMâ¦",extractCss:"Extracting CSS tokensâ¦",quadReasoning:"Quad-Fusion reasoningâ¦",rebuildTsx:"Reconstructing TSXâ¦",injectCanvas:"Injecting on canvasâ¦",demoReady:"Demo clone ready â",pagesCloned:"{{count}} page(s) cloned â",failed:"Failed"},toast:{exodusDone:"{{count}} page(s) rebuilt · Plan {{plan}}. {{next}}",upgradeElite:"Upgrade to Elite (â¬28) to clone the full site.",renameBrand:"Rename the brand in the AMY panel to finalize."},voice:{cloneDone:"{{count}} pages cloned. Step 1: replace {{brand}} with your new name in the AMY panel on the right."},mode:{demo:"DEMO MODE · Blank canvas",clone:"CLONE MODE · Target | Clone | Code",create:"CREATE MODE · Preview | Code",matrix:"DEEP DARK MATRIX · {{count}} NODES · QUAD-FUSION ROUTING"},node:{frontend:"Frontend",backend:"Backend",database:"Database",ai:"Quad-Fusion AI"},nodeTitle:{fe:"App Shell",be:"Edge Function",db:"Schema",ai:"Quad-Fusion AI"},languageBadge:"Language ·"},playground:{aucune_cle_api_trouvee_visitez:"No API key found. Visit your Dashboard to generate one.",connectez_vous_pour_utiliser_l:"Sign in to use the Playground."},translationgenerator:{turkish_turkce:"Turkish (Türkçe)"},themis:{themis:{themis_automatise_l:"Themis automates"},themisdashboard:{dashboard_client_themis_depose:"Thémis client dashboard: submit your documents, launch Bates stamping, conflict check and AI summary. Complete audit trail, GDPR certificate of destruction.",erreur_inconnue:"Unknown error",themis_a_termine_l:"Themis finished the",document_supprime_certificat_d:"Document deleted. Destruction certificate issued.",okcount_document_s_chiffre_s_a:"${okCount} AES-256 encrypted document(s) + Bitcoin anchored. Keys downloaded.",echec_file_name_msg:"Failed: ${file.name} â ${msg}",erreur:"Error",impossible_de_charger_vos_docu:"Unable to load your documents"},themisverify:{erreur:"Error",verifiez_l:"Check the"},trustcenter:{journal_cryptographique_public:"Public cryptographic journal of Thémis operations: Bates stamping, conflict checks, certificates of destruction."}}},Pu={back:"Back to Home",heroBadge:"Legal AI · Trust-Grade",heroTitle:"100 % automatic, 100 % auditable eDiscovery.",heroTagline:"Thémis eDiscovery â 100% automated. 100% auditable.",heroDescription:"Thémis is a Trust-Grade legal eDiscovery platform that automates the collection, analysis, numbering and certification of your legal documents. It secures your chain of evidence with immutable Bates stamping, multi-jurisdiction conflict check, OCR + AI summary and GDPR destruction certificate â so law firms, from solo practitioners to Big Law, reduce delays, costs and compliance risks.",ctaAutomatic:"100% AUTOMATIC",pillarsLabel:"8 PILLARS",pillarsTitle:"The Trust-Grade Arsenal",flowLabel:"FLOW",benchmarkLabel:"BENCHMARK",tarifsLabel:"PRICING",pillars:{bates:{title:"Immutable Bates Stamping",desc:"Sequential numbering ORG-MATTER-000001 with SHA-256 cryptographic proof on every produced page."},conflict:{title:"Automatic Conflict Check",desc:"Multi-entity detection via Jaccard similarity + linguistic normalization. Built-in ethical walls."},juridictions:{title:"15 native jurisdictions",desc:"FR, CH, DE, US, UK, CA, AU, SG, AE, SA, JP, IT, ES, BR, NL â compliant data localization and retention."},destruction:{title:"Destruction Certificate",desc:"Timestamped SHA-256 proof for every GDPR deletion. Publicly verifiable via the Trust Center."},ocr:{title:"OCR + AI Summary",desc:"Multimodal processing via Lovable AI Gateway. Starter â Enterprise plans with transparent rate-limiting."},zeroKnowledge:{title:"Zero-Knowledge AES-256",desc:"Encryption in your browser before upload. The Themis server never sees content in clear text."},bitcoin:{title:"Bitcoin Anchored (OpenTimestamps)",desc:"Every document is anchored to the Bitcoin blockchain via OpenTimestamps. Immutable proof of existence, publicly verifiable."}
811,rfc:{title:"RFC 3161 Timestamp (FreeTSA)",desc:"Qualified cryptographic timestamp token issued by an independent Timestamping Authority. eIDAS Advanced compatible."}},plans:{free:{tokens:"1 GB · 3 documents/month",target:"Discovery",features:["Bates stamping","Bitcoin Anchor","AdES Advanced","Public verification"]},pro:{tokens:"20 GB · 100 docs/month",target:"Individual lawyer",features:["Everything Free","Zero-Knowledge AES-256","OCR + AI Summary","5 jurisdictions","Email support"]},business:{tokens:"200 GB · 500 docs/month · 30 Swisscom QES",target:"2-10 lawyer firm",features:["Everything Pro","15 jurisdictions","30 Swisscom QES signatures","Co-branded audit trail","5 users","Priority support"]},enterprise:{tokens:"Unlimited · Unlimited QES",target:"Big Law · Legal Direction",features:["Everything Business","SAML SSO + MFA","SLA 99.99%","Private API","Dedicated CSM","Extended audit logs"]}},flow:{s1:{t:"Secure upload",d:"Drag & drop · immediate AES-256 encryption."},s2:{t:"Conflict + Jurisdiction",d:"Automatic verification in < 3 sec."},s3:{t:"Bates + OCR + AI",d:"Sequential stamping + OCR + summary."},s4:{t:"Audit + Destruction",d:"SHA-256 trail + GDPR certificate."}},jur:{eu:{region:"European Union (27)",note:"Directly admissible, no transposition needed"},ch:{region:"Switzerland",note:"QES = handwritten signature equivalent"},uk:{region:"United Kingdom",note:"Full recognition post-Brexit"},us:{region:"USA",note:"Bitcoin proof = self-authenticating (FRE 901/902)"},ca:{region:"Canada",note:"Federal + provinces recognized"},au:{region:"Australia",note:"eIDAS mirror"},sg:{region:"Singapore · ð¦ðª UAE · ð¯ðµ Japan",law:"eIDAS mirror laws",note:"Direct admissibility"},world:{region:"+ 50 other jurisdictions",note:"UN international standard"}},cmp:{r1:"eIDAS AdES Signature",r2:"Native Bitcoin Anchoring",r3:"Zero-Knowledge (blind server)",r4:"Free public verification",r5:"Cryptographic Bates Stamping",r6:"Multi-jurisdiction Conflict Check",r7:"GDPR destruction certificate",r8:"Quad-Fusion AI (auto audit)",r9:"Native jurisdictions",partial:"Partial",usFirst:"US-first",frOnly:"ð«ð· only",chOnly:"ð¨ð only"},priceSuffix:{month:"/month",lifetime:"/lifetime"},cta:{quote:"Request a quote",createFree:"Create free account",start:"Get started",trustCenter:"Trust Center"},explainer:{title:"Concrete Use Cases & Proven Law-Firm Gains",intro:"Thémis is the only Trust-Grade eDiscovery platform combining cryptographic Bates stamping, Bitcoin anchoring (OpenTimestamps), multi-jurisdiction conflict check and GDPR destruction certificate â 100% auditable, admissible in 80+ countries.",usageTitle:"Real Use Cases",usages:["Big Law & M&A: immutable Bates stamping of tens of thousands of exhibits with Bitcoin-timestamped proof.","Cross-border litigation: automatic conflict check across 15 jurisdictions (EU, CH, UK, US, CA, AU, SGâ¦).","GDPR-first firms: SHA-256 signed destruction certificate opposable to regulators.","Compliance & internal audit: OCR + AI contract summarisation in Zero-Knowledge (server never sees plaintext)."],gainsTitle:"Measured Law-Firm Gains",gains:["â92% time on discovery preparation (vs manual paralegal input).","Free public Bitcoin verification: zero counter-expertise cost for opposing counsel.","eIDAS AdES by default + optional QES Swisscom â legal force equivalent to a handwritten signature (SCSE Art. 14).","Client-side AES-256-GCM encryption: GDPR / attorney-client privilege enforced by cryptography, not by contract."],revolution:"DocuSign, Yousign and Skribble sign PDFs. Thémis signs the truth: every hash is anchored on Bitcoin, publicly verifiable by any judge, without depending on us.",superiority:"Why Infernex Thémis wins: DocuSign offers no Bitcoin anchoring, no Zero-Knowledge, no GDPR destruction certificate. Yousign is France-only and lacks multi-jurisdiction conflict check. Skribble stays confined to Switzerland with no AI OCR. None of them offer free public verification. Thémis ships all four in a single platform, admissible in 80+ countries."}},Tu={title:"Quantum Hybrid Switch",masterKill:"Master Kill Switch",masterKillDesc:"OFF = classical Supabase only. ON = autonomous IonQ/D-Wave routing.",on:"ON",off:"OFF",enabled:"Enabled",disabled:"Disabled",keyOk:"key â",keyMissing:"missing",keyPending:"pending",minRoi:"Min ROI %",stagnation:"Stagnation iters",maxCost:"Max $/strike",totalCost:"Total cost",totalGain:"Total gain",recentStrikes:"Recent quantum strikes",noStrikes:"No strikes yet â predator dormant.",killOffWarning:"Kill-switch OFF: Infernex limited to classical Supabase.",saved:"Saved",loading:"Loading configâ¦",loadError:"Failed to load quantum config"},Eu={title:"Howells Decrypt â Sovereign Multichain",subtitle:"Reconstruct private keys across BTC, ETH, SOL via Quantum Annealing. Crypto-only, FIAT strictly forbidden.",fiatBlockTitle:"ABSOLUTE FILTERING PROTOCOL",fiatBlockDesc:"FIAT, bank accounts, IBAN, SWIFT, wire transfers, credit cards are systematically rejected. Crypto-native assets only.",fiatBlocked:"FIAT detected â rejected",threshold:"Asset threshold",activation:"Activation fee",successFee:"Success fee",inD3:"in D3 tokens",assetChain:"Asset chain",estimatedUsd:"Estimated value (USD)",minNotice:"min $1M",destAddress:"Destination wallet (80% of recovered funds)",destNote:"Must be on the same chain as the recovered asset",fragments:"Memory fragments (clues, patterns, context)",fragmentsPlaceholder:"Anything you remember about the wallet creation: words, dates, devices, applications, password patternsâ¦",smartContractNotice:"By submitting, you authorize the Smart Contract to be the sole arbiter of fund distribution. The decryption attempt is reviewed by our quantum operators and may take weeks. No refunds on activation fees.",startResearch:"Start Research",submitted:"Request submitted to quantum review",validationError:"Please verify all fields. Asset must be ⥠$1M.",history:"My requests",viewProof:"View Proof of Dominance â"},Ru={title:"The Untouchables, Decrypting.",subtitle:"Live registry of cryptographic puzzles undergoing entropy reduction. Mathematical certainty, not promises.",processingPower:"Current processing power",disclaimer:"All resolutions are verified by zero-knowledge proofs. We never expose private keys."},Ou={title:"Founder Treasury (Crypto-Only)",fiatBlockTitle:"FIAT STRICTLY BLOCKED",fiatBlockDesc:"No bank, no FIAT, no IBAN/SWIFT. Only crypto-native flows. Auto-swap to USDT/USDC/DAI basket.",autoSwap:"Auto-Swap to Stables",autoSwapDesc:"Convert all incoming BTC/ETH/D3 to USDT/USDC/DAI via 1inch (max 1% impact).",stablesBasket:"Stablecoins basket",maxImpact:"Max price impact",founderPriority:"Founder liquidation priority post-decrypt",integrationNote:"Integration stub. Provide 1inch API key + signing wallet to activate on-chain swaps.",on:"ON",off:"OFF",saved:"Saved",loading:"Loadingâ¦"},Mu={gas:{title:"Quantum Gas â $D3 Fuel",depleted:"FUEL DEPLETED â 402 Payment Required",active:"ACTIVE",balance:"Balance",consumed:"Consumed",burned:"Burned 90%",foun
811der:"Founder 10%",refuelUni:"Refuel â Uniswap V4",refuelRay:"Refuel â Raydium",formula:"Auto-burn 90/10 â 90% of every $D3 consumed is burned irrevocably; 10% is routed to the Sovereign Founder Treasury as stablecoin (USDT/USDC/DAI)."}},qu={header:"International Compliance â G-NOM Protocol",body:"Technological mandatary for the recovery of dormant digital assets (Art. 2272 French Civil Code, Bona Vacantia, Unclaimed Property Acts). EXCLUSIVE SCOPE: BLOCKCHAIN ASSETS ONLY. NO INTERACTION WITH TRADITIONAL BANKING SYSTEMS. Strictly private use within sovereign zones (Geneva, Zug, Dubai, Monaco, Singapore, Panama).",routing:"STRICT ROUTING: Stablecoin only (USDT/USDC/DAI) â Zero FIAT â Zero ETH storage â Certificate injected as on-chain metadata."},Du={title:"Titanium Shield v2.6 â Manual Sovereign Gateway"},Lu={header:"INFERNEX Advisor",badge:"AI · no signup",placeholder:"Describe your need â Infernex AI picks the right plan and prepares your API key.",inputPlaceholder:"Type your need hereâ¦",thinking:"Analyzingâ¦",recommended:"Recommended",generateKey:"Generate my API key",error:"The advisor is temporarily unavailable."},Fu={seo:{title:"AI Inference Pricing Comparison 2026: Infernex vs OpenAI",description:"2026 AI inference pricing comparison. Compare tokens per dollar for OpenAI, Groq, and AWS Bedrock against Infernex's 95% lower rates. Updated March 2026."},badge:"PRICING COMPARISON â MARCH 2026",heroTitlePre:"AI Inference",heroTitleHighlight:"Pricing",heroTitlePost:"Comparison",heroSubtitle:"Side-by-side pricing for every major AI inference provider. See exactly how much you can save with Infernex.",ctaCalculate:"Calculate Your Savings",ctaTryFree:"Try Free",tableTitlePre:"Price Per",tableTitleHighlight:"Million Tokens",tableSubtitle:"Updated March 2026. Output token prices.",tableModel:"Model",whyTitlePre:"Why Infernex Is the",whyTitleHighlight:"Cheapest",faqTitlePre:"Pricing",faqTitleHighlight:"FAQ",ctaTitle:"Start Saving on AI Inference Today",ctaSubtitle:"Free tier. No credit card. 100% OpenAI-compatible.",ctaButton:"Get Started Free",cards:[{title:"Optimized Infrastructure",desc:"Custom GPU scheduling and model quantization reduce compute costs by 10x."},{title:"Semantic Caching",desc:"Intelligent response caching saves tokens on repeated queries â up to 60% additional savings."},{title:"No Markup",desc:"We pass infrastructure savings directly to you. No hidden fees, no surprise charges."}],faqItems:[{question:"What is AI inference pricing?",answer:"AI inference pricing is the cost charged per token (or per request) when you use an AI model via API. Providers like OpenAI charge $5-30 per million tokens, while Infernex offers the same models from $0.05/M â up to 95% cheaper."},{question:"Why is Infernex so much cheaper?",answer:"We use optimized inference infrastructure with semantic caching, model quantization, and efficient GPU scheduling. This lets us deliver the same quality at a fraction of the cost."},{question:"How does per-token pricing work?",answer:"You pay based on the number of tokens (words/subwords) processed. Input tokens and output tokens may have different prices. Infernex simplifies this with transparent per-million-token pricing."},{question:"Is there a free tier?",answer:"Yes. Infernex includes a generous free tier with 100K tokens/month. No credit card required to start."}]},Nu={demoComplete:"Demo Complete",unlockUnlimited:"Unlock unlimited IAO for $17/mo",getIAO:"Get IAO â $17/mo",createFreeAccount:"Create Free Account",inputPlaceholder:"Try IAO â ask anythingâ¦",remainingMessages_one:"{{count}} free message remaining",remainingMessages_other:"{{count}} free messages remaining",upgradeUnlimited:"Upgrade for unlimited"},_u={popup:{badge:"Founders Club",title:"â¬59/month <accent>for life</accent>",subtitle:"Instead of the public â¬99/mo Starter price. Founder price â¬59/mo locked for life. Limited to <strong>500 seats</strong>.",remainingPrefix:"Only ",remainingSuffix:" seats left out of 500",cta:"Claim my seat",decline:"No thanks, I prefer to pay the public price later",close:"Close",termsTitle:"Founder 500 Terms",terms:{price:"â¬59/mo locked for life â never increases",limit:"500 seats max · first confirmed payments f
811irst",eligibility:"1 seat per person · new customers only"}},offer:{seoTitle:"Founders Club â â¬59/month for life | Infernex.ai",seoDesc:"500 seats only. Founder pricing â¬59/month guaranteed for life. Same as the Starter AI Workforce plan.",cancelledError:"Payment cancelled. You can retry.",limitedBadge:"Limited offer â closes forever at 500",heroTitlePrefix:"Founders ",heroTitleAccent:"Club",heroSubtitle:"â¬59/month <accent>for life</accent> instead of the public â¬99/month Starter. Locked even when public pricing increases. 500 seats, never reopened.",seatsRemaining:"Seats remaining",foundersJoined:"{{count}} founders already joined â updated in real time",benefits:{lifetime:"Lifetime locked pricing",tokens:"500k tokens/month included",priority:"Priority feature access"},soldOutTitle:"SOLD OUT",soldOutDesc:"All 500 founder seats are taken. Public pricing from â¬99/mo.",firstname:"First name",lastname:"Last name",email:"Email",submitLoading:"Creating checkout...",submit:"Lock my seat â â¬59/month for life",stripeNote:"Secure Stripe payment. Account auto-created. Cancel anytime.",errors:{sold_out:"All 500 seats are taken. The offer is closed.",already_member:"This email is already a Founders Club member.",invalid_email:"Invalid email.",missing_fields:"All fields are required.",invalid_code:"Validation error.",unexpected:"Unexpected server response."},includedTitle:"What's <accent>included for life</accent>",included:{tokens:{t:"500,000 tokens / month",d:"Gemini 2.5 Flash, Flash Lite, Pro â automatic intelligent routing."},api:{t:"OpenAI-compatible API",d:"Drop-in: change the URL, keep your SDK. 2-minute migration."},cache:{t:"Semantic cache included",d:"30-40% hits = you pay less, you respond faster."},platform:{t:"8 Platform tools",d:"GEO, Themis, MiroFish, IAO, Quantum, Photonique, WhiteLabel."},sla:{t:"Swiss Made SLA 99.99%",d:"Geneva ð¨ð infrastructure, GDPR, AES-256, Miro Fish audit."},forever:{t:"Lifetime locked price",d:"Even when public pricing goes up, you stay at â¬59/month."}},compareTitle:"Plan comparison",compare:{col:{you:"Founders (you)",starter:"Starter public",pro:"Business public"},price:"Price /month",priceYou:"â¬59 for life",tokens:"Tokens included",overage:"Overage",rate:"Rate limit",cache:"Semantic cache",platform:"8 Platform tools",guarantee:"Lifetime guaranteed price",badge:"Founder badge",seats:"Seats",seatsYou:"500 max â closed after",seatsOther:"unlimited",priceStarter:"â¬99"},savingsLine:"Annual value: <strong>â¬480</strong> vs Starter today · <strong>â¬5,280</strong> vs Business. Over 5 years: up to <strong>â¬26,400</strong>.",faqTitle:"FAQ",faq:[{q:"Is the â¬59 price really for life?",a:"Yes. As long as your subscription stays active, the price never changes â even if public pricing increases."},{q:"What happens if I cancel?",a:"You permanently lose the founder pricing. No reactivation possible: the 500 seats are closed once taken."},{q:"Can I use the API in production?",a:"Yes. 99.99% SLA, Swiss Made infrastructure, OpenAI SDK compatible. It's the exact same API as Enterprise clients."},{q:"What if I exceed the included tokens?",a:"Overage billed â¬0.20 / 1M tokens â ~95% cheaper than OpenAI. No cutoff, no surprise."},{q:"How do I get access after payment?",a:"Stripe â automatic webhook â account created â magic link emailed to you. Up and running in under 60 seconds."}],finalRemaining:"Only <strong>{{count}}</strong> seats left out of 500.",emailPh:"[email protected]",reqMin:"{{n}} req/min",overage:"{{p}} / 1M tokens",termsTitle:'"Founder 500" Terms',terms:{price:{t:"Lifetime price",d:"â¬59/month â locked for life while your subscription stays active. Never increases. Annual option â¬590/year (2 months free), also locked."},limit:{t:"500 seats maximum",d:"Allocated by confirmed payment order (Stripe timestamp). Permanently closed at seat #500 â never reopens."},refDate:{t:"Reference date",d:"Terms in force as of July 28, 2026. The live public counter is authoritative for remaining seats."},eligibility:{t:"Eligibility",d:"1 seat per person or company. New customers only (no active paying Infernex subscription). Valid email and bank card required. Personal, non-transferable offer."},cancellation:{t:"Cancellation",d:"Cancel anytime. On cancellation or non-payment: your seat and founder pricing are permanently lost."}},faq500Title:"Founder 500 FAQ",faq500:[{q:"Who is eligible for the Founder 500 offer?",a:"Any developer, freelancer or startup creating an Infernex account with a valid email address. One seat per person and per company. Accounts created solely to resell access are excluded."},{q:"Why is the offer capped at 500 seats?",a:"The 59â¬/month lifetime rate is only sustainable on a closed volume. Once the 500 seats are taken, the offer closes permanently and the public Starter plan stays at 99â¬/month."},{q:"What is the reference date of the offer?",a:"The offer opened on July 28, 2026 and closes automatically on the 500th reserved seat, without notice. Reserved seats left unpaid after 24h are released back into the pool."},{q:"What happens if I cancel my subscription?",a:"Your seat is released and the founder rate is permanently lost. No reactivation at 59⬠is possible, even if seats become available again."},{q:"Can I change plan later?",a:"Yes. You can move to Business or Enterprise at any time. Returning to the 59â¬/month Founder plan remains possible as long as your founder subscription was never cancelled."}]},admin:{title:"Founders Club members",filterPh:"Filter by email/name...",csv:"CSV",noMembers:"No members.",col:{email:"Email",first:"First name",last:"Last name",subscribed:"Joined",tokens7d:"Tokens 7d",status:"Status"},stat:{signups:"Signups",remaining:"{{n}} remaining",mrr:"MRR",mrrSub:"monthly recurring",today:"Today",thisWeek:"{{n}} this week",active7d:"Active 7d",activeSub:"active members",reqDay:"Requests / day",reqMonth:"Requests / month",tokensSub:"{{n}} tokens",cost:"AI cost / month",costSub:"Gemini API",margin:"Gross margin",marginSub:"MRR â AI cost"},status:{active:"Active",pending:"Pending",past_due:"Past due",canceled:"Canceled"},funnel:{title:"Founder 500 funnel",impressions:"Popup impressions",clicks:"Popup clicks",views:"Offer page views",checkouts:"Stripe checkouts",subs:"Subscriptions activated",activeNow:"{{n}} active",last24h:"Last 24 hours",last7d:"Last 7 days",miniSummary:"{{i}} imp · {{c}} clicks · {{ch}} checkouts · {{s}} subs",noData:"No events tracked yet â the funnel fills as soon as the first impressions come in."},lowSeats:{title:"â ï¸ Founder 500 alert â only {{n}} seats left",desc:"Fewer than 20 founder seats remaining. The offer closes automatically at 0. An email alert was sent to the admin."}}},zu={explainer:{title:"TSU PBit â Use Cases & Proven Enterprise Gains",intro:"TSU (Thermodynamic Sampling Unit) PBit is the ultimate precision engine: Langevin dynamics + Gibbs sampling on probabilistic p-bits deliver 99.7% factual accuracy â the highest verifiable precision available on the market today.",usageTitle:"Concrete Business Use Cases",usages:["Financial services: zero-drift LLM answers for risk models, compliance memos and trading desks.","Healthcare & pharma: clinical decision support and drug-interaction analysis with regulator-grade traceability.","Legal & tax: contract review, jurisprudence and regulatory Q&A where a single hallucination costs millions.","Defense & government: sovereign, auditable inference with cryptographic proof of every output."],gainsTitle:"Measured Enterprise Gains",gains:["99.7% factual precision â a 20à reduction of residual error vs GPT-4o and Claude 3.5.","Zero factual drift certified over 10M inferences (Langevin convergence, signed SHA-256).","Regulatory-ready: SOC2, EU AI Act, HIPAA, ISO 27001 evidence bundle generated on demand.","All-in-One bundle: replaces OpenAI Enterprise + Google Quantum AI + a private photonics lab."],revolution:"TSU PBit is a physics breakthrough: probabilistic bits sampled through thermodynamic dynamics converge on the true answer instead of the most likely one. Precision stops being a hope â it becomes a physical law.",superiority:"Why Infernex wins:
811no competitor ships thermodynamic p-bits at production scale. OpenAI, Anthropic and Google top out at ~96% factual accuracy. TSU PBit ships 99.7% â the only inference engine safe enough for mission-critical enterprise workloads."},page:{seoTitle:"TSU PBit Thermodynamic AI Engine",seoDescription:"Standalone TSU PBit product page for 99.7% factual precision, Langevin dynamics, Gibbs sampling and zero-drift enterprise inference.",badge:"Standalone TSU PBit Engine",title:"TSU PBit",titleHighlight:"99.7% thermodynamic precision",subtitle:"A dedicated physics-native inference engine for finance, healthcare, legal and defense workloads where a single hallucination is unacceptable.",primaryCta:"Start with TSU PBit",secondaryCta:"View precision dashboard",stats:{precisionLabel:"factual precision",residualLabel:"residual error",methodLabel:"sampling method"},cardsTitle:"Built for critical decisions",cardsSubtitle:"TSU PBit is not a bundle. It is the standalone thermodynamic layer that cleans uncertainty before your answer reaches production.",card1Title:"Thermodynamic convergence",card1Desc:"Probabilistic p-bits use controlled noise to converge toward the most reliable answer instead of amplifying model drift.",card2Title:"Bayesian-grade sampling",card2Desc:"Gibbs sampling and Langevin dynamics refine local probabilities for high-stakes reasoning and uncertainty management.",card3Title:"Audit-ready precision",card3Desc:"Every inference can be paired with a signed proof trail for compliance, governance and mission-critical review.",dashboardTitle:"TSU Precision Dashboard",dashboardSubtitle:"The standalone TSU path focuses on residual-error cleaning before deployment into regulated workflows."}},ju={seo:{title:"Infernex AI Workforce Pricing â Starter, Business, Enterprise",description:"Deploy your autonomous AI workforce. Compare Starter, Business, and Enterprise plans for Infernex AI employees."},badge:"AI WORKFORCE â PLANS",title:"Deploy your AI workforce",subtitle:"Three simple plans to scale autonomous AI employees across your business.",apiLink:"Looking for developer API pricing?",apiLinkCta:"See Infernex API plans",perMonth:"/month",popular:"Most popular",ctaStart:"Start Free Trial",ctaContact:"Contact Sales",starter:{name:"Starter",price:"â¬99",desc:"For small teams launching their first AI workforce.",f1:"3 AI Employees",f2:"Built-in CRM",f3:"Basic automations",f4:"Email + Slack connectors",f5:"Community support"},business:{name:"Business",price:"$499",desc:"For growing companies scaling AI operations.",f1:"10 AI Employees",f2:"ROI Analytics dashboard",f3:"Advanced integrations (Salesforce, HubSpot, Shopify)",f4:"AI Safety Layer + approvals",f5:"Priority support"},enterprise:{name:"Enterprise",price:"Custom",desc:"For organizations with mission-critical deployments.",f1:"Unlimited custom AI workforce",f2:"Advanced security & SSO/SAML",f3:"Private deployment (VPC / on-prem)",f4:"Dedicated success manager",f5:"99.99% SLA + audit logs"}},Gu={badge:"ENTERPRISE · 30-MIN DEMO",title:"See your AI workforce, live.",subtitle:"A 30-minute personalized session with our solutions team. We map your workflows, quantify the ROI, and show a working AI employee for your industry.",b1:"Response within 24 hours",b2:"GDPR / SOC 2 · EU data residency",b3:"Tailored to your team size & stack",b4:"Free ROI report after the call",trust:"Trusted by teams at scale-ups and enterprises across Europe, US and MENA. Swiss-made · Genève ð¨ð",thankTitle:"Request received",thankBody:"Our team will reach out within 24 hours to schedule your demo.",viewPricing:"View pricing",readBlog:"Read case studies",success:"We received your request. Our team will reach out within 24 hours.",errorWorkEmail:"Please use a corporate email address.",error:"Something went wrong. Please try again or email [email protected].",submitting:"Sendingâ¦",submit:"Book my demo",privacy:"By submitting, you agree to be contacted about Infernex products. We never share your data.",select:"Selectâ¦",messagePlaceholder:"A few sentences on your current stack and where you want AI to take over.",seo:{title:"Book a Demo â Infernex AI Workforce for Enterprise",description:"See how Infernex AI Workforce automates your operations. 30-minute personalized demo. Response in under 24 hours."},fields:{name:"Full name",email:"Work email",company:"Company",size:"Team size",useCase:"Primary use case",timeline:"Timeline",budget:"Monthly budget range",message:"What are you trying to solve? (optional)"}},Bu={marketplace:{title:"Hire your AI employee",subtitle:"Pick a role, deploy in minutes. Every template is opinionated, editable, and safe by default.",customBuilder:"Custom builder",searchPlaceholder:"Search roles, skills, toolsâ¦",all:"All",noResults:"No template matches your search.",decrease:"Decrease quantity",increase:"Increase quantity",deploying:"Deployingâ¦",quickDeployN:"Quick deploy {{count}}",customize:"Customize",deployedToast:"Deployed {{count}} {{name}} AI employees.",deployedToastSingle:"{{name}} deployed from marketplace.",signupToHire:"Sign up to hire {{name}}",quantity:"Quantity",deployTitle:"Deploy {{title}}",employeeName:"Employee name",employeeTitle:"Title",mission:"Mission",connectTools:"Connect compatible tools",connectorsHint:"Real credential wiring happens in <connectors>Connectors</connectors> after deployment.",instructions:"Instructions (system prompt)",approval:"Require human approval before external actions (recommended)",cancel:"Cancel",deployAIEmployee:"Deploy AI Employee",sizingTitle:"How many AI employees do you need?",sizingLine:"By department: Supp
811ort 5â10 · Sales 5â10 · Marketing 2â4 · Legal 1â2 · Dev 3â5 · HR/Finance 1â2. Typical total: 20â40+ agents for 50+ employees. Use the counter to hire multiple copies."},deployedStrip:{title:"Your deployed AI employees",hint:"Click an agent to chat",viewAll:"View all Workforce"}},Uu={home:"Home",all:"All",search:"Searchâ¦",deploy:"Deploy AI Employee",pricing:"See pricing",demo:"Book a demo",connect:"Connect",docs:"Docs â",viewConnector:"View â",viewEmployee:"View â",viewConnectors:"Browse Connectors",viewEmployees:"Browse AI Employees",connectorsShort:"connectors",supportedActions:"Supported actions",compatibleEmployees:"Compatible AI Employees",compatibleEmployeesEmpty:"Ready to be assigned to any custom AI Employee.",setup:"Setup guide",setupStep1:"Open the Connector Marketplace in your workspace.",setupStep2:"Authenticate via",setupStep3:"Grant the requested scopes for the actions you want to enable.",setupStep4:"Assign the connector to one or more AI Employees.",related:"Related connectors",relatedEmployees:"Related AI Employees",businessValue:"Business value",workflows:"Example workflows",supportedConnectors:"Supported connectors",roi:"Estimated ROI",faq:"Frequently asked questions",filterByConnector:"Filter by connector:",connectors:{badge:"Connector Marketplace",title:"Connect once. Reuse everywhere.",subtitle:"80+ connectors shared across every AI Employee. Authenticate once â no duplicated logins.",short:"Connectors",metaTitle:"Connector Marketplace â Infernex AI Workforce",metaDescription:"Connect Salesforce, Slack, Stripe, GitHub, Snowflake and 80+ tools once. Every authorized AI Employee reuses your connections."},employees:{badge:"AI Workforce Marketplace",title:"25+ AI Employees. One Workforce.",subtitle:"From CFO Analyst to DevOps Engineer â pre-trained, connector-aware, and ready to deploy.",short:"AI Employees",crossTitle:"Meet the AI Employees",crossDesc:"25+ AI Employees ready to use your connected tools.",metaTitle:"AI Employee Marketplace â Infernex",metaDescription:"Hire pre-trained AI Employees for finance, sales, marketing, legal, engineering, data and more. Connect once â deploy anywhere."},category:{crm:"CRM",communication:"Communication",productivity:"Productivity",development:"Development",cloud:"Cloud",databases:"Databases",erp:"ERP",finance:"Finance",marketing:"Marketing",seo:"SEO",support:"Support",ecommerce:"E-commerce",documents:"Documents",meetings:"Meetings",ai_providers:"AI Providers"},department:{finance:"Finance",sales:"Sales",marketing:"Marketing",legal:"Legal",supply_chain:"Supply Chain",engineering:"Engineering",data:"Data",hr:"HR",executive:"Executive"}},Qu={hero:{badge:"Infernex AI Workforce",pre:"Launch an AI-powered company powered by dozens of AI agents and orchestrated by one",accent:"Master AI Agent",post:"â the superbrain that designs, deploys, and manages everything for you. From idea to global-scale company.",sub:"Build a complete team of autonomous AI employees that work 24/7, <0>automate</0> complex and repetitive tasks, give your team an unprecedented level of Fortune 500-grade performance, and cut all your AI inference costs by up to 95% with intelligent LLM routing.",ctaPrimary:"Create my first AI employee â free",ctaSecondary:"Build with Infernex API",microcopy:"No credit card. 50K free tokens/month. OpenAI SDK compatible."}},Vu={title:"Developers: your first API call in 90 seconds",subtitle:"Generate a key, change one line (base_url), run your first request. No sales call.",cta:"Start in 90 seconds"},Hu={seo:{title:"Super Admin â Infernex Control Tower",desc:"Private supervisor control tower where the Master AI Agent runs Infernex itself using the existing AI Workforce."},title:"Super Admin â Infernex Control Tower",subtitle:"The existing Master AI Agent operates Infernex like a customer Enterprise company. It reuses the existing AI Workforce. Nothing executes without your approval.",nav:{commandCenter:"Master AI Command Center",rc2:"RC2 Innovation Center",integrations:"Integration Center",runPass:"Run supervisor pass"},kpi:{openAudits:"Open audits",pending:"Pending approvals",workforce:"AI Workforce",lastRun:"Last supervisor run",users:"Users",subs:"Active subscriptions",api24h:"API calls (24h)",alerts7d:"Security alerts (7d)",newSupport:"New support messages",activeSuffix:"{{active}}/{{total}} active",running:"runningâ¦"},tabs:{approvals:"Approval Center",support:"Support Inbox",health:"API Health",audits:"Continuous Audit",workforce:"AI Workforce (Infernex)",proposals:"Employee Proposals",reports:"Executive Reports",memory:"Memory",infra:"Infrastru
811cture",ops:"Ops Intelligence"},approvals:{approveAll:"Approve all pending",empty:"No items yet. Run a supervisor pass.",approve:"Approve",delay:"Delay",markExecuting:"Mark executing",done:"Done",reject:"Reject",masterReasoning:"Master reasoning:",generatedPrompt:"Generated Lovable prompt"},audits:{empty:"No findings yet."},workforce:{note:"Existing AI Employees assigned by the Master Agent to operate Infernex. Read-only â this is the same workforce used for customers."},proposals:{note:"Last-resort proposals for new AI Employees. Approval reuses the existing AI employees table.",empty:"No proposals.",mission:"Mission:",reason:"Reason:",roi:"Expected ROI:",approveAdd:"Approve & add to workforce",reject:"Reject"},reports:{placeholder:"The generated executive report will appear here."},toast:{reportReady:"Report ready: {{kind}}",runComplete:"Supervisor run complete",supervisorFailed:"Supervisor failed",failed:"Action failed",copied:"Copied",itemUpdated:"Item {{action}}"},inbox:{filters:{all:"All",new:"New",read:"Read",handled:"Handled"},refresh:"Refresh",loadFailed:"Inbox load failed",updateFailed:"Update failed",deliveryFailed:"{{count}} support email(s) failed delivery â the sender domain notify.infernex.ai is still verifying its DNS.",notQueued:"{{count}} message(s) were never queued.",nothingLost:"Nothing is lost: every message is stored and readable here.",empty:"No messages in this view.",queued:"queued",notQueuedBadge:"not queued",notePlaceholder:"Internal noteâ¦",reply:"Reply",replySubject:"Re: [{{category}}] Infernex support",copyAddress:"Copy address",emailCopied:"Email copied",markRead:"Mark as read",markHandled:"Mark handled",reopen:"Reopen"},health:{openIncidents:"{{count}} open incident(s)",since:"since {{date}} · {{count}} alert(s) sent",allHealthy:"All {{count}} monitored endpoints are healthy",failing:"{{count}} endpoint(s) failing",runCheck:"Run check now",failureDrill:"Failure drill",loading:"Loading health telemetryâ¦",noProbes:"No probe recorded in the last 24h. Run a check to seed telemetry.",up:"UP",down:"DOWN",p95:"p95 {{value}} ms",uptime24h:"24h uptime {{value}}%",probes:"{{count}} probes",last:"last {{time}}",resolved:"Resolved incidents",checkDone:"Health check completed",checkDoneDrill:"Failure drill probe injected.",checkDoneAll:"All targets probed and analyzed.",checkFailed:"Health check failed"},daily:{title:"Daily Master Prompt",subtitle:"One consolidated prompt per day covering every pending request â no more prompt-by-prompt.",generate:"Generate today's prompt",copy:"Copy",empty:"No master prompt yet. Generate the consolidated prompt for today.",covers:"Covers {{count}}
811 requests",approveAll:"Approve all covered items",markDone:"Mark as done",reject:"Reject",generated:"Master prompt generated ({{count}} requests merged)",nothingPending:"Nothing pending today",updated:"Master prompt updated",failed:"Master prompt failed"},infra:{heartbeats:"Automation heartbeats",regions:"Multi-region synthetics",avgLatency:"Avg {{value}} ms",burnRate:"Burn rate (Ada)",dailyOpex:"Daily opex ${{value}}",avg30d:"30-day avg ${{value}}",needsApproval:"Human approval required",chiefOfStaff:"AI Chief of Staff",missionCompletion:"Mission completion rate",capabilityAdherence:"Capability adherence",reprioritised:"{{count}} tasks re-prioritised",openIncidents:"{{count}} open incidents",customerHealth:"Customer health (Milo)",healthy:"{{count}} healthy",atRisk:"{{count}} at risk",criticalAccounts:"{{count}} critical",reconciliation:"Data reconciliation",noDiscrepancies:"No discrepancies detected",seoBriefs:"Automated SEO briefs",noData:"No data yet"},ops:{security:{title:"Security incident triage",pending:"pending",empty:"No security incident. All alerts triaged."},workforce:{title:"AI employee performance",employees:"Employees",anomalies:"Anomalies",gaps:"Skill gaps",reportDate:"Report",missing:"Missing capabilities",noGap:"Full strategic capability coverage â no gap detected."},churn:{title:"Churn prediction",atRisk:"at risk",queue:"Proactive outreach queue"},cache:{title:"API cache",hits:"hits",empty:"Cache empty â first requests will populate it."}}},Wu={title:"Daily action plan",subtitle:"One consolidated plan per day covering every pending action of your company â approve once, the console clears itself.",generate:"Generate today's plan",copy:"Copy",copied:"Copied",empty:"No plan yet today. Generate the consolidated plan for your company.",covers:"Covers {{count}}
811 actions",approveAll:"Approve all covered actions",markDone:"Mark as done",reject:"Reject",generated:"Daily plan generated ({{count}} actions merged)",nothingPending:"Nothing pending today",updated:"Daily plan updated",failed:"Daily plan failed"},Ku={autopilot:{title:"Autopilot running",desc:"Your company is being built automatically: design, validation, AI workforce, brand and website.",done:"Autopilot complete",doneDesc:"Company deployed and website generated.",partial:"Autopilot in progress",partialDesc:"Remaining steps continue automatically in the background."},website:{viewSite:"View website",liveUrl:"Live preview URL",domainStatus:"Official domain waiting for DNS",dnsRecords:"DNS records"},swarm:{title:"Multi-source AI swarm",capacity:"{{n}} virtual agents capacity",zeroCost:"$0/month open-source orchestration",activeHeads:"Active AI employees",departments:"Departments ready",max:"Max swarm",note:"Your real employees orchestrate several free open-source agent frameworks (Moltbook, CrewAI, AG2/AutoGen, LangGraph, MetaGPT, CAMEL-AI, Swarms, OpenHands, AgentVerse, ChatDevâ¦). New frameworks are discovered automatically every day, so capacity keeps growing without any paid infrastructure.",sources:"Open-source sources"}},$u={eyebrow:"Ready-made companies",title:"Buy a company already built and operated by AI",subtitle:"Each company was designed, built, staffed and deployed by our Master AI agent. Live website, AI workforce, operational workflows, $0/month infrastructure.",singlePrice:"One single price for every company:",oneTime:"one-time",gateTitle:"The catalogue is reserved for registered members",gateBody:"Create your free account or sign in to see every available ready-made company and buy it in one click.",gateSignup:"Create my free account",gateLogin:"Sign in",count:"{{count}} companies available",sold:"Sold",available:"Available",multiIndustry:"Multi-industry",defaultSummary:"Operational AI company with a live website and a deployed AI workforce.",units:"{{n}} departments run by AI",pages:"{{n}} website pages live",zeroCost:"$0/month infrastructure",buy:"Buy this company",empty:"New companies are being published by the Master AI agent. Come back shortly.",error:"Unable to load the catalogue",checkoutError:"Checkout unavailable",seoTitle:"Ready-made AI companies for sale | Infernex",seoDesc:"Turnkey AI-operated companies built by the Infernex Master AI agent: live website, AI workforce, $0/month infrastructure. One single price."},Yu={eyebrow:"Ready-made companies",title:"AI companies already built are available for sale",sub:"Our Master AI agent has already designed, built and deployed complete companies: live website, AI workforce, operational workflows, $0/month infrastructure. All at one single identical price.",gate:"Catalogue visible for registered members only.",button:"See ready-made companies"},Zu={loading:"Loading site",defaultDescription:"AI-generated company website powered by Infernex.",notFoundTitle:"Site not found",notFoundDescription:"This generated company site is not published yet.",previewUnavailable:"Preview unavailable",notReady:"This site is not live yet.",backToBuilder:"Back to builder",navigation:"Site navigation",aiCompany:"AI company",primaryCta:"Start now",sections:"Page sections",poweredBy:"Powered by Infernex AI"},Ju={quantumDashboard:jc,quantumOnboarding:Gc,trustbar:Bc,socialProof:Uc,exitPopup:Qc,typewriter:Vc,radar:Hc,reliability:Wc,dataFlow:Kc,quicknav:$c,common:Yc,serverless:Zc,hireQuant:Jc,enterprise:Xc,quantumAccelerated:ed,openaiMigration:td,reduceAiCosts:nd,pitchDeck:ad,nav:od,hero:rd,features:id,savings:sd,testimonials:ld,pricing:cd,playground:dd,calculator:ud,faq:md,contact:pd,footer:fd,inference_pricing:hd,auth:gd,dashboard:bd,admin:yd,enterpriseForm:vd,cookies:Id,roiSection:Ad,slaSection:xd,industries:wd,moneyBackSection:Sd,quantum:kd,quantumVsQuant:Cd,whiteLabel:Pd,pocSection:Td,onboarding:Ed,photonique:Rd,blog:Od,ultimate:Md,heroTypewriter:qd,platformSeo:Dd,heroTestimonials:Ld,compare:Fd,models:Nd,trust:_d,geo:zd,flow:jd,tools:Gd,mirofish:Bd,promo:Ud,planOverview:Qd,quickStart:Vd,apiIntegration:Hd,savingsCalculator:Wd,usageAlerts:Kd,adminDashboards:$d,enterpriseCheckout:Yd,notFound:Zd,paymentSuccess:Jd,quantumTeaser:Xd,statusPage:eu,changelog:tu,chatbot:nu,partners:au,privacy:ou,iaoPage:ru,iaoWidget:iu,audit:su,blogQuadFusion:lu,benchmarkDefaultPrompt:cu,homeCta:du,supremacy:uu,benchmarkWidget:mu,institutionalFooter:pu,sovereignty:fu,i
811dentityGateway:hu,routing:gu,connectivity:bu,offline:yu,security:vu,settings:Iu,voice:Au,comparison:xu,auto:wu,home:Su,components:ku,pages:Cu,themisPage:Pu,"ep.badge":"ENTERPRISE AI INFRASTRUCTURE","ep.heroTitle1":"Cut Enterprise AI Costs","ep.heroTitle2":"by","ep.heroDesc":"Same models. Same quality. SOC 2 compliant. 99.99% SLA. Sub-10ms latency. 72-qubit quantum computing.","ep.requestDemo":"Request Enterprise Demo","ep.calculateROI":"Calculate ROI","ep.statsCost":"Cost Savings","ep.statsLatency":"Latency","ep.statsSLA":"Uptime SLA","ep.statsQubits":"Quantum Qubits","ep.benefitsTitle":"Enterprise-Grade","ep.benefitsHL":"AI Infrastructure","ep.before":"Before","ep.after":"After","ep.saved":"saved","ep.roiTitle":"Enterprise","ep.roiHL":"ROI Examples","ep.migTitle":"Enterprise","ep.migHL":"Migration Path","ep.faqTitle":"Enterprise","ep.seoDesc":"Enterprise AI inference at 95% lower cost. SOC 2 compliant, 99.99% SLA, sub-10ms latency.","ep.benefits.0.title":"95% Cost Reduction","ep.benefits.0.desc":"Cut annual AI spending from $5M+ to under $250K with drop-in API migration. No code changes.","ep.benefits.1.title":"Sub-10ms Latency","ep.benefits.1.desc":"Faster than OpenAI, Anthropic, and Google. Critical for real-time trading.","ep.benefits.2.title":"99.99% Uptime SLA","ep.benefits.2.desc":"Enterprise-grade reliability with contractual SLA.","ep.benefits.3.title":"SOC 2 & GDPR Compliant","ep.benefits.3.desc":"Data stays in Switzerland. Full audit trail, encryption at rest.","ep.benefits.4.title":"Dedicated Infrastructure","ep.benefits.4.desc":"Isolated compute for enterprise clients. Custom rate limits and priority queuing.","ep.benefits.5.title":"72-Qubit Quantum Computing","ep.benefits.5.desc":"Infernex Quantum QPU access for QAOA optimization and QRNG.","ep.cases.0.sector":"Investment Banking","ep.cases.0.metric":"Portfolio optimization latency: 200ms to 8ms","ep.cases.1.sector":"Insurance & Actuarial","ep.cases.1.metric":"Risk model execution: 45min to 12 seconds","ep.cases.2.sector":"Pharmaceutical R&D","ep.cases.2.metric":"Drug interaction analysis: 6hrs to 3 minutes","ep.cases.3.sector":"High-Frequency Trading","ep.cases.3.metric":"Signal generation latency: 50ms to under 5ms","ep.mig0t":"Discovery","ep.mig0d":"We audit your current AI stack and spending","ep.mig1t":"POC","ep.mig1d":"2-week proof of concept with your workloads","ep.mig2t":"Migration","ep.mig2d":"Gradual traffic shift with zero downtime","ep.mig3t":"Optimize","ep.mig3d":"Quantum-optimized inference at 95% lower cost","ep.faq.0.q":"How much can enterprise customers save?","ep.faq.0.a":"Enterprise customers typically save 90-97% on AI inference costs.","ep.faq.1.q":"Is there a minimum commitment?","ep.faq.1.a":"Custom pricing based on volume. Flexible monthly billing. No long-term lock-in.","ep.faq.2.q":"How does the migration work?","ep.faq.2.a":"Our team handles migration end-to-end. 2-4 weeks. Zero downtime guaranteed.","ep.faq.3.q":"What about data security?","ep.faq.3.a":"SOC 2 Type II, GDPR, data hosted in Switzerland. Zero data retention.","ep.faq.4.q":"Do you offer quantum computing?","ep.faq.4.a":"Yes. 72-qubit processor for QAOA and QRNG. Unique to Infernex.","qai.badge":"QUANTUM-ACCELERATED INFERENCE","qai.statsSpeedup":"Speedup","qai.statsLatency":"Latency","qai.statsPrice":"/month","qai.capTitle":"Quantum","qai.capHL":"Capabilities","qai.thClassical":"Classical","qai.thQuantum":"Quantum (72Q)","qai.thSpeedup":"Speedup","qai.ucTitle":"Industry","qai.ucHL":"Use Cases","qai.faster":"faster","qai.faqTitle":"Quantum","qai.feat.0.t":"72 Superconducting Qubits","qai.feat.0.d":"Infernex Quantum processor accessible via API.","qai.feat.1.t":"QAOA Optimization","qai.feat.1.d":"Quantum Approximate Optimization for portfolio, scheduling, combinatorial problems.","qai.feat.2.t":"QRNG Random Generation","qai.feat.2.d":"True quantum random numbers for Monte Carlo, cryptography, A/B testing.","qai.feat.3.t":"Hybrid Classical-Quantum","qai.feat.3.d":"Seamless fallback to classical algorithms if QPU unavailable.","qai.feat.4.t":"Sub-50ms Decision Latency","qai.feat.4.d":"Quantum-accelerated decisions faster than any human analyst.","qai.feat.5.t":"Enterprise-Grade Security","qai.feat.5.d":"SOC 2, GDPR. Quantum key distribution ready. Swiss-hosted.","qai.uc.0.s":"Finance & Banking","qai.uc.0.t":"Portfolio Optimization","qai.uc.0.d":"QAOA finds optimal asset allocation across 1000+ securities simultaneously.","qai.uc.1.s":"High-Frequency Trading","qai.uc.1.t":"Signal Generation","qai.uc.1.d":"Quantum random sampling eliminates predictable patterns in trading signals.","qai.uc.2.s":"Pharmaceutical R&D","qai.uc.2.t":"Molecular Simulation","qai.uc.2.d":"Quantum-assisted molecular dynamics for drug interaction modeling.","qai.uc.3.s":"Insurance & Actuarial","qai.uc.3.t":"Risk Modeling","qai.uc.3.d":"QRNG-powered Monte Carlo simulations for VaR, CVaR, stress testing.","qai.bm.0.m":"QAOA Optimization","qai.bm.0.c":"45 min","qai.bm.0.q":"12 sec","qai.bm.1.m":"Monte Carlo (1M paths)","qai.bm.1.c":"8 min","qai.bm.1.q":"0.3 sec","qai.bm.2.m":"Portfolio Rebalancing","qai.bm.2.c":"2 hours","qai.bm.2.q":"4 sec","qai.bm.3.m":"Random Key Generation","qai.bm.3.c":"Pseudo-random","qai.bm.3.q":"True QRNG","qai.bm.4.m":"A/B Test Assignment","qai.bm.4.c":"Math.random()","qai.bm.4.q":"72-qubit QRNG","qai.faq.0.q":"What is quantum-accelerated AI inference?","qai.faq.0.a":"Combines classical AI with quantum computing for optimization (QAOA) and randomness (QRNG).","qai.faq.1.q":"How many qubits does Infernex use?","qai.faq.1.a":"72 superconducting qubits, one of the most powerful processors via cloud API.","qai.faq.2.q":"Do I need quantum expertise?","qai.faq.2.a":"No. Our API abstracts the quantum layer. Standard REST endpoints.","qai.faq.3.q":"What if QPU is unavailable?","qai.faq.3.a":"Automatic classical fallback. Your applications never stop working.","qai.faq.4.q":"Is this real quantum or simulation?","qai.faq.4.a":"Real quantum hardware. 72-qubit superconducting processor.","auto.25eff4ce31":"Infernex","auto.e89d951207":"INFERNEX","auto.7fe4952363":"AI Infrastru
811cture","auto.09d18199e4":"San Francisco · Dubai · Geneva","auto.8a30376276":"The Problem","auto.dcbed5cd61":"Killing","auto.32f28a2c73":"Innovation","auto.080b92196c":"per 1M tokens (GPT-4/Claude)","auto.b0efbdd9af":"6 mo","auto.9aafc2dc24":"client =","auto.4d8283b751":"apiKey:","auto.ca05e152dd":'"sk-..."',"auto.90f9efc4b3":'"inf_sk_..."',"auto.19bb6bf78f":"baseURL:","auto.9b62305de1":"Intelligent","auto.2e948be420":"Model Selection","auto.87861f20d8":"Simple queries","auto.c136ea4a94":"â Llama 3 / Mistral","auto.43dbed7663":"Code generation","auto.2a61c3e4be":"â Code-specialized","auto.347dd7bdfa":"Complex reasoning","auto.dc35481d15":"Core Product","auto.471f395113":"Enterprise-Grade","auto.4da121a391":"AI Inference","auto.2cabf8233a":"Product Suite","auto.ca8a2087e5":"Three","auto.738ea6dd30":"Revenue Engines","auto.ae2630cd44":"MiroFish","auto.d58e45b419":"API & CI/CD integration","auto.b157e0ffcf":"From $49/month","auto.95e53de21c":"$12B","auto.cabcc5d084":"GEO Platform","auto.9e53dce8d4":"From $17/month","auto.5031f4a81d":"$80B","auto.6f1933d17f":"Quantum Computing","auto.ce33d690b3":"Quantum A/B testing","auto.67fe389c99":"From $499/month","auto.958b35a5e4":"$65B","auto.26901eb12d":"Model Ecosystem","auto.77a29b9003":"Global Reach","auto.5fcb9c43aa":"Languages supported","auto.efe01da942":"RTL","auto.d88946b678":"SEO","auto.176d74bf6b":"Market Opportunity","auto.ce4660e033":"$300B+ Combined","auto.5c08a08813":"TAM","auto.57e832e6ca":"$150B","auto.839d2c22fb":"GEO/SEO","auto.033b1e7914":"Quantum","auto.5319adef79":"Traction","auto.e41a4815b5":"What's Already","auto.955ad3298d":"Live","auto.92d235dff1":"ð Go-to-Market","auto.e1533dc293":"â
Referral/affiliate program","auto.cb912f2c6b":"Business Model","auto.6c0de86049":"Multi-Product","auto.54358a914f":"Revenue","auto.d0f88e519e":"/mo","auto.4c608a3f64":"MiroFish SaaS","auto.0940fb4908":"Quantum Add-on","auto.02e9ce0986":"Competitive Landscape","auto.5e5d8619da":"Why Infernex","auto.f4e8734a42":"Wins","auto.21021ea0e5":"Feature","auto.0523b13262":"OpenAI","auto.747168b2e7":"Groq","auto.95d83e986c":"Together","auto.ac848fa228":"Team","auto.d6b258792f":"Organization","auto.1d36783e12":"Chart","auto.2741df6932":"Director & Founder","auto.b59862101b":"CFO","auto.ad47487ba9":"Hiring","auto.981acafcff":"CTO","auto.029714fc86":"Engineers Team","auto.27c7152b07":"The Ask","auto.8d703e4ebb":"Raising a","auto.dba005190e":"Seed","auto.b7f41fc141":"Round","auto.2bb4147c6b":"18-Month Milestones","auto.29afbc227d":"Roadmap","auto.3fb1cbff01":"What's","auto.10ac3d0425":"Next","auto.66c8d2593d":"www.infernex.ai","auto.0fdda3e908":"[email protected]","auto.c4f8c8924a":"Smart Routing","auto.8ebcf77d4a":"AI analyzes complexity & routes to best model","auto.b06f676c46":"< 5ms Overhead","auto.926fd358ea":"Routing adds negligible latency","auto.bdcd7957d4":"OpenAI Compatible","auto.d05d349a4c":"Change 2 lines of code. That's it.","auto.289a70abfa":"15 Languages","auto.ea944861c0":"Full platform localization for global reach","auto.7d9faab09c":"API Request","auto.1a4dc90a18":"OpenAI-compatible","auto.8d5a63bdeb":"AI Classifier","auto.b1f29308b1":"Complexity analysis","auto.02da81ab8e":"Smart Router","auto.52593d8e8a":"Optimal model pick","auto.d64ed3e9c1":"Response","auto.7103dd12e1":"< 50ms total","auto.a70d5f644a":"Works with LangChain, LlamaIndex, any OpenAI integration","auto.9a6e5f2010":"Full feature parity with OpenAI API","auto.c4f3aa1c5c":"Multi-provider failover, automatic fallbacks","auto.447247757c":"Live Playground","auto.afe9d86a08":"Test any model in-browser before integration","auto.8f0dc4dde6":"White-Label API","auto.601adfd51d":"Partners resell under their own brand","auto.d1944e9c65":"30-Day Money-Back Guarantee","auto.b69d92a99d":"Risk-free for all plans","auto.94cb3d601b":"Hugging Face","auto.e80bd6b371":"Google AI","auto.a2fe9051fe":"NVIDIA","auto.539134de8f":"Mistral","auto.83113c1ec0":"50+ models","auto.935ee74f4d":"Auto-selection","auto.6fa1ff0ee3":"95% savings","auto.3752601bfb":"<5ms","auto.ee6085db6c":"Starter","auto.672caf27f5":"Developer","auto.abd900517e":"Pro","auto.a0f2763826":"Enterprise","auto.60f74b8dc6":"Vice CEO","auto.b690a03de4":"Marketing Team","auto.cc4b6be328":"Support Team","auto.5240352e61":"Engineering & Infrastru
811cture","auto.ca7e9d389a":"Go-to-Market & Sales","auto.85837304e0":"Product (GEO + Quantum + MiroFish)","auto.5f00c0e07e":"Operations & Reserve","d3WorldChallenges.badge":"SPECIAL OPERATIONS","d3WorldChallenges.title":"World Challenges â War Room","d3WorldChallenges.viral":"Time is no longer an obstacle. D3CODE breaks the chains of the past. Watch the truth set itself free, live. Next stop: total sovereignty. #D3CODE #Singularity","d3WorldChallenges.targets":"Active targets","d3WorldChallenges.bounty":"Total bounty","d3WorldChallenges.burned":"$D3 burned","d3WorldChallenges.unit":"Unit","d3WorldChallenges.tracker":"Live Tracker","d3WorldChallenges.progress":"Decryption","d3WorldChallenges.bountyShort":"Bounty","d3WorldChallenges.burnedShort":"Burned","d3WorldChallenges.terminal":"Quad-Fusion Terminal","d3WorldChallenges.waiting":"// initializing Quad-Fusion engineâ¦","d3WorldChallenges.burnNote":"Each attempt consumes and burns $D3, increasing scarcity with every cycle.","nav.tagWarRoom":"War Room",quantumSwitch:Tu,howells:Eu,proof:Ru,treasury:Ou,d3:Mu,gnom:qu,titanium:Du,serviceAdvisor:Lu,aiPricing:Fu,iaoDemo:Nu,founders:_u,tsu:zu,workforcePricing:ju,bookDemo:Gu,aiWorkforce:Bu,marketplace:Uu,wh:Qu,platformQuickstart:Vu,superadmin:Hu,clientDaily:Wu,companyBuilder:Ku,rm:$u,rmCta:Yu,publicSite:Zu},Ve={en:{nav:{features:"Features",savings:"Savings",testimonials:"Testimonials",pricing:"Pricing",faq:"FAQ",language:"Language",login:"Login",signup:"Sign up",products:"Products",closeMenu:"Close menu",openMenu:"Open menu",support:"Support",blog:"Blog",developers:"Developers",workforce:"AI Workforce",hire:"Hire",workforcePricing:"Pricing",tagAutonomous:"Autonomous employees",tagAgents12:"12 AI templates",tagStarter99:"From â¬59/mo",tagLegalAI:"Legal AI",inference:"Inference API",tagOpenAI:"OpenAI-compatible",tagBuild:"Build with Infernex"},institutionalFooter:{badgeLabels:{gdpr:"GDPR Compliant",tls13:"TLS 1.3 Encryption",swissHosted:"Swiss-hosted Infrastructure",rlsAuth:"Row-level Security"}},wh:{seo:{title:"Infernex AI â Deploy Your Autonomous AI Workforce",desc:"Infernex AI Workforce connects your data, tools and processes to deploy AI employees that work 24/7 with your teams â powered by Infernex Core infrastructure."},hero:{badge:"Infernex AI Workforce",line1:"Hire AI employees and create your company in",line1Accent:"one click",line2:"From an idea to a future Fortune 500 company.",moreInfo:"More info",ctaPrimary:"Create my first AI employee â free",ctaSecondary:"Build with Infernex API",microcopy:"No credit card. 50K free tokens/month. OpenAI SDK compatible.",valueProp:"Companies lose millions every year to repetitive work. Infernex deploys autonomous AI employees that automate your business 24/7."},emp:{sales:{n:"AI Sales Agent",d:"Qualifies inbound leads, personalizes outreach and updates your CRM in real time."},support:{n:"AI Customer Support Agent",d:"Resolves L1 tickets 24/7, escalates edge cases, connects to your helpdesk."},marketing:{n:"AI Marketing Agent",d:"Plans campaigns, drafts multi-channel copy, tracks performance and iterates."},finance:{n:"AI Finance Agent",d:"Reconciles invoices, forecasts cash flow and flags anomalies before they cost you."},research:{n:"AI Research Agent",d:"Scans sources, synthesizes briefs and keeps your team ahead of the market."},legal:{n:"AI Legal Agent â powered by Themis",d:"Themis-governed AI paralegal: reviews NDAs, DPAs and MSAs under confidentiality, GDPR and full audit-trail rules."}},eng:{ph:"Photonic inference for sub-millisecond latency.",q:"72-qubit quantum routing for combinatorial workloads.",t:"Thermodynamic sampling, 99.7% precision.",m:"500,000 distributed agents for parallel reasoning."},val:{tasks:{k:"Tasks automated",v:"Repetitive workflows handed off to autonomous agents."},time:{k:"Time reclaimed",v:"Your team focuses on decisions, not data entry."},cost:{k:"Cost reduction",v:"Infernex Core routes to the cheapest engine that meets your SLA."},roi:{k:"Measurable ROI",v:"Every action is logged, auditable and tied to a business outcome."}},s1:{eyebrow:"What is AI Workforce",title:"AI employees, not just AI tools.",body:"Each Infernex AI employee has a defined role, its own instructions, access to your tools, and long-term memory of your business.",f1:{t:"A role, not a prompt",d:"Each employee is scoped to a business function â with objectives, KPIs and guardrails."},f2:{t:"Connected to your stack",d:"CRM, helpdesk, email, docs, databases. Actions are executed, not just suggested."},f3:{t:"Governed and auditable",d:"Every action logged. Role-based access. GDPR-compliant. Swiss engineered."}},s2:{eyebrow:"In minutes, not months",title:"Create an AI employee in three steps.",st1:{t:"Pick a role",d:"Sales, Support, Marketing, Finance, Research â or a custom role."},st2:{t:"Connect your tools",d:"One-click connectors to your CRM, helpdesk, docs and databases."},st3:{t:"Deploy and iterate",d:"Your employee starts working. Review actions, adju
811st guardrails, scale."},cta:"Try the employee builder"},s3:{eyebrow:"Ready-to-deploy roles",title:"Meet your AI teammates.",explore:"Explore role"},s4:{eyebrow:"Why teams choose Infernex",title:"Measurable outcomes, not vanity metrics.",body:"We do not publish invented benchmarks. The actual ROI depends on your workflows and volume."},s5:{eyebrow:"The engine underneath",title:"Powered by Infernex Core.",body:"Every AI employee runs on Infernex Core â an intelligent router that dispatches each task to photonic, quantum, thermodynamic or MiroFish engines.",cta:"See the developer platform"},s6:{eyebrow:"Workforce vs the old way",title:"One platform. Not ten disconnected tools.",old:{h:"Classic stack",t:"Multiple software, silos, brittle automations.",i1:"A separate SaaS for every function.",i2:"Zapier chains that break when APIs change.",i3:"Standalone AI tools with no memory of your business.",i4:"Costs stack up. Governance gets lost."},new:{h:"Infernex AI Workforce",t:"One workforce, one platform, one bill.",i1:"Autonomous AI employees per business function.",i2:"Unified connectors to your CRM, docs and helpdesk.",i3:"Long-term memory of your data and processes.",i4:"Auditable actions, role-based governance, one bill."},cta:"Create my first AI employee",microcopy:"Free forever plan. Upgrade any time."},faq:{eyebrow:"Frequently asked questions",title:"Everything you need to know",subtitle:"Deploy, govern and scale your AI workforce with confidence.",q1:"What is an AI employee, exactly?",a1:"An AI employee is an autonomous agent with a defined role, instructions, tool permissions and long-term memory of your business.",q2:"How long does it take to deploy the first AI employee?",a2:"Under 5 minutes. Pick a template, connect your data source, review the mission and permissions, and deploy.",q3:"Is my company data safe and GDPR-ready?",a3:"Yes. Every action is logged, isolated per organization and protected by human approval for sensitive operations.",q4:"Do I keep human control over my AI employees?",a4:"Always. You define permissions per tool, per action and per data source.",q5:"Which tools and CRMs can my AI employees connect to?",a5:"Salesforce, HubSpot, Zoho, Pipedrive, Slack, Gmail, Google Drive, Notion, database, Stripe, Zendesk and 30+ more.",q6:"How does pricing work?",a6:"Start free with limited usage. Paid plans unlock more employees, tokens and connectors.",q7:"Can I use my own LLM or route to a specific model?",a7:"Yes. Infernex Core routes each task to the most efficient engine or to your chosen provider.",q8:"What ROI can I realistically expect?",a8:"Teams typically report a 40â70% reduction in repetitive workload within the first 60 days.",cta:"Talk to our team"}}},fr:{nav:{features:"Fonctionnalités",savings:"Ãconomies",testimonials:"Témoignages",pricing:"Tarifs",faq:"FAQ",language:"Langue",login:"Connexion",signup:"S'inscrire",products:"Produits",closeMenu:"Fermer le menu",openMenu:"Ouvrir le menu",support:"Support",blog:"Blog",developers:"Développeurs",workforce:"Workforce IA",hire:"Recruter",workforcePricing:"Tarifs",tagAutonomous:"Employés autonomes",tagAgents12:"12 modèles IA",tagStarter99:"Dès 59â¬/mois",tagLegalAI:"IA juridique",inference:"API dâinférence",tagOpenAI:"Compatible OpenAI",tagBuild:"Construire avec Infernex"},institutionalFooter:{badgeLabels:{gdpr:"Conforme RGPD",tls13:"Chiffrement TLS 1.3",swissHosted:"Infrastructure hébergée en Suisse",rlsAuth:"Sécurité par ligne"}},wh:{seo:{title:"Infernex AI â Déployez votre équipe IA autonome",desc:"Infernex AI Workforce connecte vos données, outils et processus pour déployer des employés IA qui travaillent 24/7 avec vos équipes."},hero:{badge:"Infernex Workforce IA",line1:"Recrutez des employés IA et créez votre entreprise en",line1Accent:"un clic",line2:"D'une idée à une future entreprise du Fortune 500.",moreInfo:"Plus d'infos",ctaPrimary:"Créer mon premier employé IA â gratuit",ctaSecondary:"Construire avec l'API Infernex",microcopy:"Sans carte bancaire. 50K jetons gratuits/mois. Compatible avec le SDK OpenAI.",valueProp:"Les entreprises perdent des millions chaque année en travail répétitif. Infernex déploie des employés IA autonomes qui automatisent votre business 24h/24."},emp:{sales:{n:"Agent IA commercial",d:"Qualifie les leads entrants, personnalise les relances et met à jour votre CRM en temps réel."},support:{n:"Agent IA support client",d:"Résout les tickets L1 24/7, escalade les cas sensibles et se connecte à votre centre de support."},marketing:{n:"Agent IA marketing",d:"Planifie les campagnes, rédige les contenus multicanaux, suit la performance et itère."},finance:{n:"Agent IA finance",d:"Rapproche les factures, prévoit la trésorerie et signale les anomalies avant quâelles ne coûtent."},research:{n:"Agent IA recherche",d:"Analyse les sources, synthétise les briefs et garde votre équipe en avance sur le marché."},legal:{n:"Agent IA juridique â propulsé par Themis",d:"Assistant juridique IA gouverné par Themis : révise NDA, DPA et MSA dans le respect de la confidentialité, du RGPD et dâun audit complet."}},eng:{ph:"Inférence photonique pour une latence inférieure à la milliseconde.",q:"Routage quantique 72 qubits pour les charges comb
811inatoires.",t:"Ãchantillonnage thermodynamique, précision 99,7 %.",m:"500 000 agents distribués pour le raisonnement parallèle."},val:{tasks:{k:"Tâches automatisées",v:"Les processus répétitifs sont confiés à des agents autonomes."},time:{k:"Temps récupéré",v:"Votre équipe se concentre sur les décisions, pas sur la saisie."},cost:{k:"Réduction des coûts",v:"Infernex Core route vers le moteur le moins cher qui respecte votre SLA."},roi:{k:"ROI mesurable",v:"Chaque action est journalisée, auditable et reliée à un résultat métier."}},s1:{eyebrow:"Quâest-ce que la force de travail IA",title:"Des employés IA, pas seulement des outils IA.",body:"Chaque employé IA Infernex possède un rôle défini, ses propres instructions, lâaccès à vos outils et une mémoire long terme de votre entreprise.",f1:{t:"Un rôle, pas un prompt",d:"Chaque employé est cadré sur une fonction métier â avec objectifs, KPI et garde-fous."},f2:{t:"Connecté à vos outils",d:"CRM, centre de support, e-mail, documents, bases de données. Les actions sont exécutées, pas seulement suggérées."},f3:{t:"Gouverné et auditable",d:"Chaque action est journalisée. Accès par rôle. Conforme RGPD. Ingénierie suisse."}},s2:{eyebrow:"En minutes, pas en mois",title:"Créez un employé IA en trois étapes.",st1:{t:"Choisir un rôle",d:"Ventes, Support, Marketing, Finance, Recherche â ou un rôle sur mesure."},st2:{t:"Connecter vos outils",d:"Connecteurs en un clic vers votre CRM, centre de support, documents et bases de données."},st3:{t:"Déployer et itérer",d:"Votre employé commence à travailler. Vérifiez les actions, ajustez les garde-fous, puis montez en charge."},cta:"Essayer le créateur dâemployé"},s3:{eyebrow:"Rôles prêts à déployer",title:"Rencontrez vos coéquipiers IA.",explore:"Explorer le rôle"},s4:{eyebrow:"Pourquoi les équipes choisissent Infernex",title:"Des résultats mesurables, pas des métriques de vanité.",body:"Nous ne publions pas de tests comparatifs inventés. Le retour sur investissement réel dépend de vos processus et volumes."},s5:{eyebrow:"Le moteur sous-jacent",title:"Propulsé par Infernex Core.",body:"Chaque employé IA fonctionne sur Infernex Core â un routeur intelligent qui envoie chaque tâche vers les moteurs photonique, quantique, thermodynamique ou MiroFish.",cta:"Voir la plateforme développeur"},s6:{eyebrow:"Force de travail IA vs lâancienne méthode",title:"Une plateforme. Pas dix outils déconnectés.",old:{h:"Suite classique",t:"Logiciels multiples, silos, automatisations fragiles.",i1:"Un SaaS séparé pour chaque fonction.",i2:"Des chaînes Zapier qui cassent quand les API changent.",i3:"Des outils IA isolés sans mémoire de votre entreprise.",i4:"Les coûts sâempilent. La gouvernance disparaît."},new:{h:"Infernex Workforce IA",t:"Une force de travail IA, une plateforme, une facture.",i1:"Employés IA autonomes par fonction métier.",i2:"Connecteurs unifiés vers votre CRM, vos documents et votre centre de support.",i3:"Mémoire long terme de vos données et processus.",i4:"Actions auditables, gouvernance par rôle, une seule facture."},cta:"Créer mon premier employé IA",microcopy:"Plan gratuit à vie. Passez à un plan supérieur à tout moment."},faq:{eyebrow:"Questions fréquentes",title:"Tout ce que vous devez savoir",subtitle:"Déployez, gouvernez et faites évoluer votre force de travail IA en confiance.",q1:"Quâest-ce quâun employé IA exactement ?",a1:"Un employé IA est un agent autonome avec un rôle défini, ses propres instructions, des permissions sur vos outils et une mémoire long terme de votre entreprise.",q2:"Combien de temps faut-il pour déployer le premier employé IA ?",a2:"Moins de 5 minutes. Choisissez un modèle, connectez votre source de données, vérifiez la mission et les permissions, puis déployez.",q3:"Les données de mon entreprise sont-elles sécurisées et prêtes RGPD ?",a3:"Oui. Chaque action est journalisée, isolée par organisation et protégée par une validation humaine pour les opérations sensibles.",q4:"Est-ce que je garde le contrôle humain sur mes employés IA ?",a4:"Toujours. V
811ous définissez les permissions par outil, action et source de données.",q5:"à quels outils et CRM mes employés IA peuvent-ils se connecter ?",a5:"Salesforce, HubSpot, Zoho, Pipedrive, Slack, Gmail, Google Drive, Notion, base de données, Stripe, Zendesk et plus de 30 autres.",q6:"Comment fonctionne la tarification ?",a6:"Démarrez gratuitement avec un usage limité. Les plans payants débloquent plus dâemployés, tokens et connecteurs.",q7:"Puis-je utiliser mon propre LLM ou router vers un modèle précis ?",a7:"Oui. Infernex Core route chaque tâche vers le moteur le plus efficace ou vers le fournisseur choisi.",q8:"Quel ROI puis-je attendre de façon réaliste ?",a8:"Les équipes constatent généralement 40 à 70 % de réduction de charge répétitive dans les 60 premiers jours.",cta:"Parler à notre équipe"}}}},Xu={es:{nav:{features:"Funcionalidades",savings:"Ahorros",testimonials:"Testimonios",pricing:"Precios",faq:"FAQ",language:"Idioma",login:"Iniciar sesión",signup:"Registrarse",products:"Productos",closeMenu:"Cerrar menú",openMenu:"Abrir menú",support:"Soporte",blog:"Blog",developers:"Desarrolladores",workforce:"Fuerza laboral IA",hire:"Contratar",workforcePricing:"Precios",tagAutonomous:"Empleados autónomos",tagAgents12:"12 plantillas IA",tagStarter99:"Desde 59â¬/mes",tagLegalAI:"IA legal",inference:"API de inferencia",tagOpenAI:"Compatible con OpenAI",tagBuild:"Construir con Infernex"},institutionalFooter:{badgeLabels:{gdpr:"Conforme GDPR",tls13:"Cifrado TLS 1.3",swissHosted:"Infraestructura alojada en Suiza",rlsAuth:"Seguridad por fila"}},wh:{seo:{title:"Infernex AI â Despliega tu fuerza laboral IA autónoma",desc:"Infernex AI Workforce conecta tus datos, herramientas y procesos para desplegar empleados IA que trabajan 24/7 con tus equipos."},hero:{badge:"Infernex Fuerza laboral IA",line1:"Contrata empleados IA y crea tu empresa en",line1Accent:"un clic",line2:"De una idea a una futura empresa Fortune 500.",moreInfo:"Más información",ctaPrimary:"Crear mi primer empleado IA â gratis",ctaSecondary:"Construir con la API Infernex",microcopy:"Sin tarjeta bancaria. 50K tokens gratis/mes. Compatible con OpenAI SDK.",valueProp:"Las empresas pierden millones cada año en trabajo repetitivo. Infernex despliega empleados IA autónomos que automatizan tu negocio 24/7."},emp:{sales:{n:"Agente IA de ventas",d:"Cualifica leads entrantes, personaliza contactos y actualiza tu CRM en tiempo real."},support:{n:"Agente IA de soporte",d:"Resuelve tickets L1 24/7, escala casos sensibles y se conecta a tu centro de soporte."},marketing:{n:"Agente IA de marketing",d:"Planifica campañas, redacta contenido multicanal, mide el rendimiento e itera."},finance:{n:"Agente IA de finanzas",d:"Concilia facturas, prevé la tesorerÃa y detecta anomalÃas antes de que cuesten dinero."},research:{n:"Agente IA de investigación",d:"Analiza fuentes, sintetiza briefings y mantiene a tu equipo por delante del mercado."},legal:{n:"Agente IA legal â impulsado por Themis",d:"Asistente legal IA gobernado por Themis: revisa NDAs, DPAs y MSAs con confidencialidad, GDPR y auditorÃa completa."}},eng:{ph:"Inferencia fotónica con latencia inferior a un milisegundo.",q:"Enrutamiento cuántico de 72 qubits para cargas combinatorias.",t:"Muestreo termodinámico, precisión del 99,7 %.",m:"500 000 agentes distribuidos para razonamiento paralelo."},val:{tasks:{k:"Tareas automatizadas",v:"Los flujos repetitivos pasan a agentes autónomos."},time:{k:"Tiempo recuperado",v:"Tu equipo se centra en decisiones, no en introducir datos."},cost:{k:"Reducción de costes",v:"Infernex Core enruta al motor más barato que cumpla tu SLA."},roi:{k:"ROI medible",v:"Cada acción queda registrada, auditable y ligada a un resultado de negocio."}},s1:{eyebrow:"Qué es la fuerza laboral IA",title:"Empleados IA, no solo herramientas IA.",body:"Cada empleado IA de Infernex tiene un rol definido, sus propias instrucciones, acceso a tus herramientas y memoria a largo plazo de tu negocio.",f1:{t:"Un rol, no un prompt",d:"Cada empleado se limita a una función de negocio â con objetivos, KPIs y barandillas."},f2:{t:"Conectado a tu stack",d:"CRM, soporte, email, documentos, bases de datos. Las acciones se ejecutan, no solo se sugieren."}
811,f3:{t:"Gobernado y auditable",d:"Cada acción registrada. Acceso por rol. Conforme GDPR. IngenierÃa suiza."}},s2:{eyebrow:"En minutos, no en meses",title:"Crea un empleado IA en tres pasos.",st1:{t:"Elige un rol",d:"Ventas, Soporte, Marketing, Finanzas, Investigación â o un rol a medida."},st2:{t:"Conecta tus herramientas",d:"Conectores de un clic a tu CRM, soporte, documentos y bases de datos."},st3:{t:"Despliega e itera",d:"Tu empleado empieza a trabajar. Revisa acciones, ajusta barandillas, escala."},cta:"Probar el creador de empleados"},s3:{eyebrow:"Roles listos para desplegar",title:"Conoce a tus compañeros IA.",explore:"Explorar rol"},s4:{eyebrow:"Por qué eligen Infernex",title:"Resultados medibles, no métricas de vanidad.",body:"No publicamos benchmarks inventados. El ROI real depende de tus procesos y volúmenes."},s5:{eyebrow:"El motor debajo",title:"Impulsado por Infernex Core.",body:"Cada empleado IA funciona sobre Infernex Core â un router inteligente que envÃa cada tarea al motor fotónico, cuántico, termodinámico o MiroFish.",cta:"Ver la plataforma para desarrolladores"},s6:{eyebrow:"Workforce vs el modelo antiguo",title:"Una plataforma. No diez herramientas desconectadas.",old:{h:"Stack clásico",t:"Múltiples softwares, silos, automatizaciones frágiles.",i1:"Un SaaS distinto para cada función.",i2:"Cadenas Zapier que se rompen cuando cambian las APIs.",i3:"Herramientas IA aisladas sin memoria de tu negocio.",i4:"Los costes se acumulan. La gobernanza se pierde."},new:{h:"Infernex Fuerza laboral IA",t:"Una fuerza laboral, una plataforma, una factura.",i1:"Empleados IA autónomos por función de negocio.",i2:"Conectores unificados a tu CRM, docs y soporte.",i3:"Memoria a largo plazo de tus datos y procesos.",i4:"Acciones auditables, gobernanza por rol, una factura."},cta:"Crear mi primer empleado IA",microcopy:"Plan gratuito para siempre. Sube de plan cuando quieras."},faq:{eyebrow:"Preguntas frecuentes",title:"Todo lo que necesitas saber",subtitle:"Despliega, gobierna y escala tu fuerza laboral IA con confianza.",q1:"¿Qué es exactamente un empleado IA?",a1:"Un empleado IA es un agente autónomo con rol definido, instrucciones, permisos sobre tus herramientas y memoria a largo plazo de tu negocio.",q2:"¿Cuánto se tarda en desplegar el primer empleado IA?",a2:"Menos de 5 minutos. Elige una plantilla, conecta tu fuente de datos, revisa la misión y los permisos, y despliega.",q3:"¿Están mis datos seguros y listos para GDPR?",a3:"SÃ. Cada acción se registra, está aislada por organización y protegida por aprobación humana en operaciones sensibles.",q4:"¿Mantengo el control humano sobre mis empleados IA?",a4:"Siempre. Defines permisos por herramienta, acción y fuente de datos.",q5:"¿A qué herramientas y CRMs se conectan?",a5:"Salesforce, HubSpot, Zoho, Pipedrive, Slack, Gmail, Google Drive, Notion, bases de datos, Stripe, Zendesk y 30+ más.",q6:"¿Cómo funciona el precio?",a6:"Empieza gratis con uso limitado. Los planes de pago desbloquean más empleados, tokens y conectores.",q7:"¿Puedo usar mi propio LLM o enrutar a un modelo concreto?",a7:"SÃ. Infernex Core enruta cada tarea al motor más eficiente o al proveedor que elijas.",q8:"¿Qué ROI puedo esperar de forma realista?",a8:"Los equipos suelen reportar una reducción del 40â70 % de la carga repetitiva en los primeros 60 dÃas.",cta:"Hablar con nuestro equipo"}}},de:{nav:{features:"Funktionen",savings:"Ersparnisse",testimonials:"Referenzen",pricing:"Preise",faq:"FAQ",language:"Sprache",login:"Anmelden",signup:"Registrieren",products:"Produkte",closeMenu:"Menü schlieÃen",openMenu:"Menü öffnen",support:"Support",blog:"Blog",developers:"Entwickler",workforce:"KI-Belegschaft",hire:"Einstellen",workforcePricing:"Preise",tagAutonomous:"Autonome Mitarbeiter",tagAgents12:"12 KI-Vorlagen",tagStarter99:"Ab 59 â¬/Monat",tagLegalAI:"Juristische KI",inference:"Inferenz-API",tagOpenAI:"OpenAI-kompatibel",tagBuild:"Mit Infernex bauen"},institutionalFooter:{badgeLabels:{gdpr:"DSGVO-konform",tls13:"TLS-1.3-Verschlüsselung",swissHosted:"In der Schweiz gehostete Infrastruktur",rlsAuth:"Sicherheit auf Zeilenebene"}},wh:{seo:{title:"Infernex AI â Setzen Sie Ihre autonome KI-Belegschaft ein",desc:"Infernex AI Workforce verbindet Ihre Daten, Tools und Prozesse und stellt KI-Mitarbeiter bereit, die rund um die Uhr mit Ihren Teams arbeiten."},hero:{badge:"Infernex KI-Belegschaft",line1:"Stellen Sie KI-Mitarbeiter ein und gründen Sie Ihr Unternehmen mit",line1Accent:"einem Klick",line2:"Von einer Idee zu einem künftigen Fortune-500-Unternehmen.",moreInfo:"Mehr erfahren",ctaPrimary:"Meinen ersten KI-Mitarbeiter erstellen â kostenlos",ctaSecondary:"Mit der Infernex API bauen",microcopy:"Keine Kreditkarte. 50K kostenlose Tokens/Monat. OpenAI-SDK-kompatibel.",valueProp:"Unternehmen verlieren jedes Jahr Millionen durch repetitive Arbeit. Infernex setzt autonome KI-Mitarbeiter ein, die Ihr Geschäft 24/7 automatisieren."},emp:{sales:{n:"KI-Vertriebsagent",d:"Qualifiziert eingehende Leads, personalisiert Ansprache und aktualisiert Ihr CRM in Echtzeit."},support:{n:"KI-Support-Agent",d:"Löst L1-Tickets rund um die Uhr, eskaliert Grenzfälle und verbindet sich mit Ihrem Helpdesk."},marketing:{n:"KI-Marketing-Agent",d:"Plant Kampagnen, verfasst Multichannel-Texte, misst die Performance und iteriert."},finance:{n:"KI-Finanz-Agent",d:"Gleicht Rechnungen ab, prognostiziert Cashflow und meldet Anomalien, bevor sie Kosten verursachen."},research:{n:"KI-Research-Agent",d:"Durchsucht Quellen, fasst Briefings zusammen und hält Ihr Team dem Markt voraus."},legal:{n:"KI-Legal-Agent â betrieben von Themis",d:"Themis-gesteuerter KI-Legal-Assistent: prüft NDAs, DPAs und MSAs unter
811Vertraulichkeit, DSGVO und vollständigem Audit-Trail."}},eng:{ph:"Photonische Inferenz mit Sub-Millisekunden-Latenz.",q:"Quanten-Routing mit 72 Qubits für kombinatorische Workloads.",t:"Thermodynamisches Sampling, 99,7 % Präzision.",m:"500.000 verteilte Agenten für paralleles Reasoning."},val:{tasks:{k:"Automatisierte Aufgaben",v:"Repetitive Workflows werden an autonome Agenten übergeben."},time:{k:"Zurückgewonnene Zeit",v:"Ihr Team konzentriert sich auf Entscheidungen, nicht auf Dateneingabe."},cost:{k:"Kostenreduktion",v:"Infernex Core routet zur günstigsten Engine, die Ihr SLA erfüllt."},roi:{k:"Messbarer ROI",v:"Jede Aktion ist protokolliert, prüfbar und an ein Geschäftsergebnis gebunden."}},s1:{eyebrow:"Was ist KI-Belegschaft",title:"KI-Mitarbeiter, nicht nur KI-Tools.",body:"Jeder Infernex-KI-Mitarbeiter hat eine definierte Rolle, eigene Anweisungen, Zugriff auf Ihre Tools und Langzeitgedächtnis Ihres Unternehmens.",f1:{t:"Eine Rolle, kein Prompt",d:"Jeder Mitarbeiter ist auf eine Geschäftsfunktion zugeschnitten â mit Zielen, KPIs und Leitplanken."},f2:{t:"Mit Ihrem Stack verbunden",d:"CRM, Helpdesk, E-Mail, Dokumente, Datenbanken. Aktionen werden ausgeführt, nicht nur vorgeschlagen."},f3:{t:"Gesteuert und prüfbar",d:"Jede Aktion protokolliert. Rollenbasierte Zugriffe. DSGVO-konform. Swiss Engineering."}},s2:{eyebrow:"In Minuten, nicht Monaten",title:"Erstellen Sie einen KI-Mitarbeiter in drei Schritten.",st1:{t:"Rolle wählen",d:"Vertrieb, Support, Marketing, Finanzen, Research â oder eine benutzerdefinierte Rolle."},st2:{t:"Tools verbinden",d:"Ein-Klick-Konnektoren zu CRM, Helpdesk, Dokumenten und Datenbanken."},st3:{t:"Bereitstellen und iterieren",d:"Ihr Mitarbeiter beginnt zu arbeiten. Aktionen prüfen, Leitplanken anpassen, skalieren."},cta:"Employee-Builder ausprobieren"},s3:{eyebrow:"Sofort einsatzbereite Rollen",title:"Lernen Sie Ihre KI-Kollegen kennen.",explore:"Rolle erkunden"},s4:{eyebrow:"Warum Teams Infernex wählen",title:"Messbare Ergebnisse, keine Vanity-Metriken.",body:"Wir veröffentlichen keine erfundenen Benchmarks. Der reale ROI hängt von Ihren Workflows und Volumina ab."},s5:{eyebrow:"Der Motor darunter",title:"Angetrieben von Infernex Core.",body:"Jeder KI-Mitarbeiter läuft auf Infernex Core â einem intelligenten Router, der jede Aufgabe an die photonische, Quanten-, thermodynamische oder MiroFish-Engine schickt.",cta:"Zur Entwicklerplattform"},s6:{eyebrow:"Workforce vs. der alte Weg",title:"Eine Plattform. Keine zehn unverbundenen Tools.",old:{h:"Klassischer Stack",t:"Viele Softwares, Silos, brüchige Automationen.",i1:"Ein separates SaaS pro Funktion.",i2:"Zapier-Ketten, die brechen, wenn sich APIs ändern.",i3:"Isolierte KI-Tools ohne Gedächtnis Ihres Unternehmens.",i4:"Kosten stapeln sich. Governance geht verloren."},new:{h:"Infernex KI-Belegschaft",t:"Eine Belegschaft, eine Plattform, eine Rechnung.",i1:"Autonome KI-Mitarbeiter pro Geschäftsfunktion.",i2:"Vereinheitlichte Konnektoren zu CRM, Docs und Helpdesk.",i3:"Langzeitgedächtnis Ihrer Daten und Prozesse.",i4:"Prüfbare Aktionen, rollenbasierte Governance, eine Rechnung."},cta:"Meinen ersten KI-Mitarbeiter erstellen",microcopy:"Für immer kostenloser Plan. Jederzeit upgraden."},faq:{eyebrow:"Häufig gestellte Fragen",title:"Alles, was Sie wissen müssen",subtitle:"Setzen Sie Ihre KI-Belegschaft mit Vertrauen ein, steuern und skalieren Sie sie.",q1:"Was ist ein KI-Mitarbeiter genau?",a1:"Ein KI-Mitarbeiter ist ein autonomer Agent mit definierter Rolle, Anweisungen, Tool-Berechtigungen und Langzeitgedächtnis Ihres Unternehmens.",q2:"Wie lange dauert das Deployment des ersten KI-Mitarbeiters?",a2:"Unter 5 Minuten. Vorlage wählen, Datenquelle verbinden, Mission und Berechtigungen prüfen, und bereitstellen.",q3:"Sind meine Firmendaten sicher und DSGVO-bereit?",a3:"Ja. Jede Aktion wird protokolliert, pro Organisation isoliert und durch menschliche Freigabe bei sensiblen Operationen geschützt.",q4:"Behalte ich die menschliche Kontrolle über meine KI-Mitarbeiter?",a4:"Immer. Sie definieren Berechtigungen pro Tool, Aktion und Datenquelle.",q5:"Welche Tools und CRMs können angebunden werden?",a5:"Salesforce, HubSpot, Zoho, Pipedrive, Slack, Gmail, Google Drive, Notion, Datenbanken, Stripe, Zendesk und 30+ weitere.",q6:"Wie funktioniert die Preisgestaltung?",a6:"Starten Sie kostenlos mit begrenzter Nutzung. Bezahlpläne schalten mehr Mitarbeiter, Tokens und Konnektoren frei.",q7:"Kann ich mein eigenes LLM verwenden oder zu einem bestimmten Modell routen?",a7:"Ja. Infernex Core routet jede Aufgabe an die effizienteste Engine oder Ihren gewählten Anbieter.",q8:"Welchen ROI kann ich realistisch erwarten?",a8:"Teams berichten typischerweise von einer Reduktion der repetitiven Arbeitslast um 40â70 % in den ersten 60 Tagen.",cta:"Mit unserem Team sprechen"}}},pt:{nav:{features:"Funcionalidades",savings:"Economia",testimonials:"Depoimentos",pricing:"Preços",faq:"FAQ",language:"Idioma",login:"Entrar",signup:"Cadastrar",products:"Produtos",closeMenu:"Fechar menu",openMenu:"Abrir menu",support:"Suporte",blog:"Blog",developers:"Desenvolvedores",workforce:"Força de trabalho IA",hire:"Contratar",workforcePricing:"Preços",tagAutonomous:"Funcionários autônomos",tagAgents12:"12 modelos de IA",tagStarter99:"A partir de 59â¬/mês",tagLegalAI:"IA jurÃdica",inference:"API de inferência",tagOpenAI:"CompatÃvel com OpenAI",tagBuild:"Construir com Infernex"},institutionalFooter:{badgeLabels:{gdpr:"Conforme GDPR",tls13:"Criptografia TLS 1.3",swissHosted:"Infraestrutura hospedada na SuÃça",rlsAuth:"Segurança por linha"}},wh:{seo:{title:"Infernex AI â Implante sua força de trabalho IA autônoma",desc:"Infernex AI Workforce conecta seus dados, ferramentas e processos para implantar funcionários IA que trabalham 24/7 com suas equipes."},hero:{badge:"Infernex Força de trabalho IA",line1:"Contrate funcionários IA e crie a sua empresa em",line1Accent:"um clique",line2:"De uma ideia a uma futura empresa Fortune 500.",moreInfo:"Mais informações",ctaPrimary:"Criar meu primeiro funcionário IA â grátis",ctaSecondary:"Construir com a API Infernex",microcopy:"Sem cartão de crédito. 50K tokens grátis/mês. CompatÃvel com OpenAI SDK.",valueProp:"As empresas perdem milhões todos os anos com trabalho repetitivo. Infernex implanta funcionários IA autônomos que automatizam seu negócio 24/7."},emp:{sales:{n:"Agente IA de vendas",d:"Qualifica leads recebidos, personaliza o contato e atualiza seu CRM em tempo real."},support:{n:"Agente IA de suporte",d:"Resolve tickets L1 24/7, escala casos sensÃveis e se conecta ao seu helpdesk."},marketing:{n:"Agente IA de marketing",d:"Planeja campanhas, redige conteúdo multicanal, mede performance e itera."},finance:{n:"Agente IA financeiro",d:"Concilia faturas, prevê fluxo de caixa e sinaliza anomalias antes que gerem custo."},research:{n:"Agente IA de pesquisa",d:"Varre fontes, sintetiza briefings e mantém sua equipe à frente do mercado."},legal:{n:"Agente IA jurÃdico â impulsionado por Themis",d:"Assistente jurÃdico IA governado por Themis: revisa NDAs, DPAs e MSAs sob confidencialidade, GDPR e trilha de auditoria completa."}},eng:{ph:"Inferência fotônica com latência abaixo do milissegundo.",q:"Roteamento quântico de 72 qubits para cargas combinatórias.",t:"Amostragem termodinâmica, precisão de 99,7 %.",m:"500 000 agentes distribuÃdos para raciocÃnio paralelo."},val:{tasks:{k:"Tarefas automatizadas",v:"Fluxos repetitivos entregues a agentes autônomos."},time:{k:"Tempo recuperado",v:"Sua equipe foca em decisões, não em digitação."},cost:{k:"Redução de custos",v:"Infernex Core roteia para o motor mais barato que atende ao seu SLA."},roi:{k:"ROI mensurável",v:"Toda ação é registrada, auditável e ligada a um resultado de negócio."}},s1:{eyebrow:"O que é força de trabalho IA",title:"Funcionários IA, não apenas ferramentas IA.",body:"Cada funcionário IA da Infernex tem uma função definida, instruções próprias, acesso à s suas ferramentas e memória de longo prazo do seu negócio.",f1:{t:"Um papel, não um prompt",d:"Ca
811da funcionário é enquadrado em uma função de negócio â com objetivos, KPIs e limites."},f2:{t:"Conectado ao seu stack",d:"CRM, helpdesk, e-mail, docs, bancos de dados. Ações são executadas, não apenas sugeridas."},f3:{t:"Governado e auditável",d:"Toda ação registrada. Acesso por papel. Conforme GDPR. Engenharia suÃça."}},s2:{eyebrow:"Em minutos, não meses",title:"Crie um funcionário IA em três passos.",st1:{t:"Escolha um papel",d:"Vendas, Suporte, Marketing, Financeiro, Pesquisa â ou um papel sob medida."},st2:{t:"Conecte suas ferramentas",d:"Conectores em um clique para CRM, helpdesk, docs e bancos."},st3:{t:"Implante e itere",d:"Seu funcionário começa a trabalhar. Revise ações, ajuste limites, escale."},cta:"Testar o criador de funcionário"},s3:{eyebrow:"Papéis prontos para implantar",title:"Conheça seus colegas IA.",explore:"Explorar papel"},s4:{eyebrow:"Por que equipes escolhem a Infernex",title:"Resultados mensuráveis, não métricas de vaidade.",body:"Não publicamos benchmarks inventados. O ROI real depende dos seus processos e volumes."},s5:{eyebrow:"O motor por baixo",title:"Impulsionado por Infernex Core.",body:"Cada funcionário IA roda em Infernex Core â um roteador inteligente que envia cada tarefa aos motores fotônico, quântico, termodinâmico ou MiroFish.",cta:"Ver a plataforma para desenvolvedores"},s6:{eyebrow:"Workforce vs o jeito antigo",title:"Uma plataforma. Não dez ferramentas desconectadas.",old:{h:"Stack clássico",t:"Vários softwares, silos, automações frágeis.",i1:"Um SaaS separado para cada função.",i2:"Cadeias Zapier que quebram quando APIs mudam.",i3:"Ferramentas IA isoladas sem memória do seu negócio.",i4:"Custos se acumulam. Governança se perde."},new:{h:"Infernex Força de trabalho IA",t:"Uma força de trabalho, uma plataforma, uma fatura.",i1:"Funcionários IA autônomos por função de negócio.",i2:"Conectores unificados para CRM, docs e helpdesk.",i3:"Memória de longo prazo dos seus dados e processos.",i4:"Ações auditáveis, governança por papel, uma fatura."},cta:"Criar meu primeiro funcionário IA",microcopy:"Plano gratuito para sempre. Faça upgrade quando quiser."},faq:{eyebrow:"Perguntas frequentes",title:"Tudo o que você precisa saber",subtitle:"Implante, governe e escale sua força de trabalho IA com confiança.",q1:"O que é exatamente um funcionário IA?",a1:"Um funcionário IA é um agente autônomo com papel definido, instruções, permissões sobre ferramentas e memória de longo prazo do seu negócio.",q2:"Quanto tempo leva para implantar o primeiro?",a2:"Menos de 5 minutos. Escolha um modelo, conecte a fonte de dados, revise a missão e as permissões, e implante.",q3:"Meus dados estão seguros e prontos para GDPR?",a3:"Sim. Toda ação é registrada, isolada por organização e protegida por aprovação humana em operações sensÃveis.",q4:"Mantenho o controle humano sobre os funcionários IA?",a4:"Sempre. Você define permissões por ferramenta, ação e fonte de dados.",q5:"A quais ferramentas e CRMs se conectam?",a5:"Salesforce, HubSpot, Zoho, Pipedrive, Slack, Gmail, Google Drive, Notion, bancos, Stripe, Zendesk e mais de 30.",q6:"Como funciona o preço?",a6:"Comece grátis com uso limitado. Planos pagos liberam mais funcionários, tokens e conectores.",q7:"Posso usar meu próprio LLM ou rotear para um modelo especÃfico?",a7:"Sim. Infernex Core roteia cada tarefa para o motor mais eficiente ou para o provedor escolhido.",q8:"Que ROI posso esperar de forma realista?",a8:"Equipes normalmente relatam redução de 40â70 % da carga repetitiva nos primeiros 60 dias.",cta:"Falar com nossa equipe"}}},zh:{nav:{features:"åè½",savings:"èç",testimonials:"客æ·è¯ä»·",pricing:"ä»·æ ¼",faq:"常è§é®é¢",language:"è¯è¨",login:"ç»å½",signup:"注å",products:"产å",closeMenu:"å
³éèå",openMenu:"æå¼èå",support:"æ¯æ",blog:"å客",developers:"å¼åè
",workforce:"AI åå·¥å¢é",hire:"éç¨",workforcePricing:"ä»·æ ¼",tagAutonomous:"èªä¸»åå·¥",tagAgents12:"12 个 AI 模æ¿",tagStarter99:"æ¯æ 59 欧å
èµ·",tagLegalAI:"æ³å¾ AI",inference:"æ¨ç API",tagOpenAI:"å
¼å®¹ OpenAI",tagBuild:"ä½¿ç¨ Infernex æå»º"},institutionalFooter:{badgeLabels:{gdpr:"符å GDPR",tls13:"TLS 1.3 å å¯",swissHosted:"ç士æç®¡åºç¡è®¾æ½",rlsAuth:"è¡çº§å®å
¨"}},wh:{seo:{title:"Infernex AI â é¨ç½²ä½ çèªä¸» AI åå·¥å¢é",de
811sc:"Infernex AI Workforce è¿æ¥ä½ çæ°æ®ãå·¥å
·åæµç¨ï¼é¨ç½² 24/7 ä¸ä½ å¢éåä½ç AI åå·¥ã"},hero:{badge:"Infernex AI åå·¥å¢é",line1:"éä½£ AI åå·¥ï¼",line1Accent:"ä¸é®å建æ¨çå
¬å¸",line2:"ä»ä¸ä¸ªæ³æ³å°æªæ¥çä¸ç 500 强ä¼ä¸ã",moreInfo:"äºè§£æ´å¤",ctaPrimary:"å
è´¹å建æç第ä¸ä¸ª AI åå·¥",ctaSecondary:"ä½¿ç¨ Infernex API æå»º",microcopy:"æ éä¿¡ç¨å¡ãæ¯æ 50K å
è´¹ tokensãå
¼å®¹ OpenAI SDKã",valueProp:"ä¼ä¸æ¯å¹´å é夿§å·¥ä½æå¤±æ°ç¾ä¸ãInfernex é¨ç½²èªä¸» AI åå·¥ï¼å
¨å¤©å 24/7 èªå¨åæ¨çä¸å¡ã"},emp:{sales:{n:"AI éå®ä»£ç",d:"çéå
¥ç«çº¿ç´¢ï¼ä¸ªæ§åå¤èï¼å¹¶å®æ¶æ´æ°ä½ ç CRMã"},support:{n:"AI 客æä»£ç",d:"å
¨å¤©åå¤ç L1 å·¥åï¼åçº§è¾¹ç¼æ¡ä¾ï¼å¯¹æ¥ä½ çå·¥åç³»ç»ã"},marketing:{n:"AI è¥é代ç",d:"è§åæ´»å¨ï¼æ°å夿¸ éææ¡ï¼è·è¸ªè¡¨ç°å¹¶æç»è¿ä»£ã"},finance:{n:"AI è´¢å¡ä»£ç",d:"对账å票ï¼é¢æµç°éæµï¼å¨é ææå¤±åæ è®°å¼å¸¸ã"},research:{n:"AI ç 究代ç",d:"æ«æèµæï¼æ±æ»ç®æ¥ï¼è®©å¢éé¢å
å¸åºã"},legal:{n:"AI æ³å¡ä»£ç â ç± Themis 驱å¨",d:"ç± Themis æ²»çç AI æ³å¡å©çï¼å¨ä¿å¯ãGDPR å宿´å®¡è®¡è·è¸ªä¸å®¡é
NDAãDPA ä¸ MSAã"}},eng:{ph:"å
忍çï¼äºæ¯«ç§çº§å»¶è¿ã",q:"72 é忝ç¹è·¯ç±ï¼å¤çç»ååå·¥ä½è´è½½ã",t:"çåå¦éæ ·ï¼ç²¾åº¦ 99.7%ã",m:"50 ä¸åå¸å¼ä»£çå¹¶è¡æ¨çã"},val:{tasks:{k:"èªå¨åä»»å¡",v:"æé夿µç¨äº¤ç»èªä¸»ä»£çã"},time:{k:"éæ¾çæ¶é´",v:"让å¢é䏿³¨äºå³çï¼èéå½å
¥ã"},cost:{k:"ææ¬éä½",v:"Infernex Core è·¯ç±å°æ»¡è¶³ SLA çæä¾¿å®å¼æã"},roi:{k:"å¯è¡¡éç ROI",v:"æ¯ä¸ªå¨ä½é½è¢«è®°å½ãå¯å®¡è®¡ï¼å¹¶ç»å®ä¸å¡ææã"}},s1:{eyebrow:"ä»ä¹æ¯ AI åå·¥å¢é",title:"AI åå·¥ï¼èä¸ä»
ä»
æ¯ AI å·¥å
·ã",body:"æ¯ä½ Infernex AI å工齿æç¡®è§è²ãèªå·±çæä»¤ãå¯¹ä½ å·¥å
·çè®¿é®æéï¼ä»¥å对ä¸å¡çé¿æè®°å¿ã",f1:{t:"ä¸ä¸ªè§è²ï¼èéæç¤ºè¯",d:"æ¯ä½åå·¥èç¦ä¸ä¸ªä¸å¡èè½ââå¸¦ç®æ ãKPI 忤æ ã"},f2:{t:"æ¥å
¥ä½ çææ¯æ ",d:"CRMãå·¥åãé®ä»¶ãææ¡£ãæ°æ®åºãè¡å¨è¢«æ§è¡ï¼èéä»
被建议ã"},f3:{t:"åæ²»çä¸å¯å®¡è®¡",d:"ææå¨ä½ççãåºäºè§è²çè®¿é®æ§å¶ã符å GDPRãç士工ç¨ã"}},s2:{eyebrow:"åªéå åéï¼èéå 个æ",title:"䏿¥å建ä¸ä½ AI åå·¥ã",st1:{t:"éæ©è§è²",d:"éå®ã客æãè¥éãè´¢å¡ãç ç©¶ââæèªå®ä¹è§è²ã"},st2:{t:"è¿æ¥ä½ çå·¥å
·",d:"ä¸é®è¿æ¥å¨æ¥å
¥ CRMãå·¥åãææ¡£ä¸æ°æ®åºã"},st3:{t:"é¨ç½²å¹¶è¿ä»£",d:"åå·¥å¼å§å·¥ä½ãæ£æ¥å¨ä½ï¼è°æ´æ¤æ ï¼ç¶åæ©å±ã"},cta:"è¯ç¨åå·¥æå»ºå¨"},s3:{eyebrow:"å¼ç®±å³ç¨çè§è²",title:"认è¯ä½ ç AI éåã",explore:"æ¢ç´¢è§è²"},s4:{eyebrow:"å¢é为ä½éæ© Infernex",title:"å¯è¡¡éçç»æï¼èéèè£ææ ã",body:"æä»¬ä¸åå¸ç¼é çåºåãçå® ROI åå³äºä½ çæµç¨ä¸ä½éã"},s5:{eyebrow:"åºå±å¼æ",title:"ç± Infernex Core 驱å¨ã",body:"æ¯ä½ AI åå·¥é½è¿è¡å¨ Infernex Core ä¸ââä¸ä¸ªæºè½è·¯ç±å¨ï¼å°æ¯ä¸ªä»»å¡åæ´¾å°å
åãéåãçå妿 MiroFish 弿ã",cta:"æ¥çå¼åè
å¹³å°"},s6:{eyebrow:"Workforce vs æ§æ¹å¼",title:"ä¸ä¸ªå¹³å°ãè䏿¯å个å²è£çå·¥å
·ã",old:{h:"ä¼ ç»æ ",t:"å¤ä¸ªè½¯ä»¶ãçå±ãèå¼±çèªå¨åã",i1:"æ¯ä¸ªèè½ä¸ä¸ªç¬ç« SaaSã",i2:"API ä¸åå°±æç Zapier 龿¡ã",i3:"对ä¸å¡æ è®°å¿çå¤ç« AI å·¥å
·ã",i4:"ææ¬å å ï¼æ²»ç丧失ã"},new:{h:"Infernex AI åå·¥å¢é",t:"ä¸ä¸ªåå·¥å¢éãä¸ä¸ªå¹³å°ãä¸å¼ è´¦åã",i1:"æä¸å¡èè½çèªä¸» AI åå·¥ã",i2:"é¢å CRMãææ¡£ä¸å·¥åçç»ä¸è¿æ¥å¨ã",i3:"å¯¹ä½ æ°æ®ä¸æµç¨çé¿æè®°å¿ã",i4:"å¯å®¡è®¡çå¨ä½ãåºäºè§è²çæ²»çãä¸å¼ è´¦åã"},cta:"å建æç第ä¸ä¸ª AI åå·¥",microcopy:"æ°¸ä¹
å
费计åãå¯éæ¶å级ã"},faq:{eyebrow:"常è§é®é¢",title:"ä½ éè¦ç¥éçä¸å",subtitle:"èªä¿¡å°é¨ç½²ãæ²»çå¹¶æ©å±ä½ ç AI åå·¥å¢éã",q1:"AI åå·¥å°åºæ¯ä»ä¹ï¼",a1:"AI åå·¥æ¯ä¸ä¸ªèªä¸»ä»£çï¼æ¥ææç¡®è§è²ãæä»¤ã对工å
·çæéï¼ä»¥å对ä¸å¡çé¿æè®°å¿ã",q2:"é¨ç½²ç¬¬ä¸ä¸ª AI åå·¥éè¦å¤ä¹
ï¼",a2:"ä¸å° 5 åéã鿍¡æ¿ï¼è¿æ¥æ°æ®æºï¼æ£æ¥ä»»å¡ä¸æéï¼ç¶åé¨ç½²ã",q3:"æå
¬å¸çæ°æ®å®å
¨ä¸ç¬¦å GDPR åï¼",a3:"æ¯çãæ¯ä¸ªå¨ä½é½è¢«è®°å½ãæç»ç»éç¦»ï¼æææä½é人工审æ¹ã",q4:"æå¯¹ AI åå·¥æ¯å¦å§ç»ä¿æäººä¸ºæ§å¶ï¼",a4:"å§ç»å¦æ¤ãä½ æå·¥å
·ãå¨ä½åæ°æ®æºå®ä¹æéã",q5:"AI åå·¥å¯ä»¥è¿æ¥åªäºå·¥å
·å CRMï¼",a5:"SalesforceãHubSpotãZohoãPipedriveãSlackãGmailãGoogle DriveãNotionãæ°æ®åºãStripeãZendesk 以å 30 å¤ä¸ªã",q6:"å®ä»·å¦ä½ï¼",a6:"å
è´¹å¼å§ï¼ç¨éæéãä»è´¹è®¡åè§£éæ´å¤åå·¥ãtokens ä¸è¿æ¥å¨ã",q7:"æè½ä½¿ç¨èªå·±ç LLM æè·¯ç±å°ç¹å®æ¨¡ååï¼",a7:"å¯ä»¥ãInfernex Core 伿æ¯ä¸ªä»»å¡è·¯ç±å°æé«æçå¼æï¼æä½ éæ©çä¾åºåã",q8:"æè½å®é
æå¾
ææ ·ç ROIï¼",a8:"å¢éé常å¨å 60 天å
çå°é夿§å·¥ä½éåå° 40â70%ã",cta:"èç³»æä»¬çå¢é"}}},ja:{nav:{features:"æ©è½",savings:"ç¯ç´",testimonials:"ã客æ§ã®å£°",pricing:"æé",faq:"ãããã質å",language:"è¨èª",login:"ãã°ã¤ã³",signup:"ç»é²",products:"製å",closeMenu:"ã¡ãã¥ã¼ãéãã",openMenu:"ã¡ãã¥ã¼ãéã",support:"ãµãã¼ã",blog:"ããã°",developers:"éçºè
",workforce:"AIã¯ã¼ã¯ãã©ã¼ã¹",hire:"æ¡ç¨",workforcePricing:"æé",tagAutonomous:"èªå¾å徿¥å¡",tagAgents12:"12åã®AIãã³ãã¬ã¼ã",tagStarter99:"æé¡59â¬ãã",tagLegalAI:"ãªã¼ã¬ã«AI",inference:"æ¨è«API",tagOpenAI:"OpenAIäºæ",tagBuild:"Infernexã§æ§ç¯"},institutionalFooter:{badgeLabels:{gdpr:"GDPRæºæ ",tls13:"TLS 1.3æå·å",swissHosted:"ã¹ã¤ã¹ãã¹ãåºç¤",rlsAuth:"è¡ã¬ãã«ã»ãã¥ãªãã£"}},wh:{seo:{title:"Infernex AI â èªå¾åAIã¯ã¼ã¯ãã©ã¼ã¹ãå±é",desc:"Infernex AI Workforceã¯ããã¼ã¿ã»ãã¼ã«ã»ããã»ã¹ãæ¥ç¶ãã24æé365æ¥ãã¼ã ã¨å
±ã«åãAI徿¥å¡ãå±éãã¾ãã"},hero:{badge:"Infernex AIã¯ã¼ã¯ãã©ã¼ã¹",line1:"AI徿¥å¡ãéãã",line1Accent:"ã¯ã³ã¯ãªãã¯ã§ä¼ç¤¾ãè¨ç«",line2:"ã¢ã¤ãã¢ããæªæ¥ã®Fortune 500伿¥ã¸ã",moreInfo:"詳ããè¦ã",ctaPrimary:"æåã®AI徿¥å¡ã使 â ç¡æ",ctaSecondary:"Infernex APIã§æ§ç¯",microcopy:"ã¯ã¬ã¸ããã«ã¼ãä¸è¦ãæ¯æ50Kç¡æãã¼ã¯ã³ãOpenAI SDKäºæã",valueProp:"伿¥ã¯æ¯å¹´ãç¹°ãè¿ã使¥ã§æ°ç¾ä¸ã失ã
811£ã¦ãã¾ããInfernex ã¯èªå¾å AI 徿¥å¡ãå±éãããã¸ãã¹ã 24 æé 365 æ¥èªååãã¾ãã"},emp:{sales:{n:"AIå¶æ¥ã¨ã¼ã¸ã§ã³ã",d:"ã¤ã³ãã¦ã³ããªã¼ããé¸å¥ãããã¼ã½ãã©ã¤ãºããã¢ããã¼ãã¨CRMã®ãªã¢ã«ã¿ã¤ã æ´æ°ãè¡ãã¾ãã"},support:{n:"AIã«ã¹ã¿ãã¼ãµãã¼ã",d:"L1ãã±ããã24/7ã§è§£æ±ºãã¨ãã¸ã±ã¼ã¹ãã¨ã¹ã«ã¬ã¼ã·ã§ã³ãããã«ããã¹ã¯ã¨é£æºãã¾ãã"},marketing:{n:"AIãã¼ã±ãã£ã³ã°",d:"ãã£ã³ãã¼ã³ãè¨ç»ãããã«ããã£ãã«ã®å稿ãå·çããææã追跡ãã¦æ¹åãã¾ãã"},finance:{n:"AI財åã¨ã¼ã¸ã§ã³ã",d:"è«æ±ã®ç
§åããã£ãã·ã¥ããã¼äºæ¸¬ãã³ã¹ãååã®ç°å¸¸æ¤ç¥ãè¡ãã¾ãã"},research:{n:"AIãªãµã¼ãã¨ã¼ã¸ã§ã³ã",d:"æ
å ±æºãã¹ãã£ã³ããããªã¼ããè¦ç´ãããã¼ã ãå¸å ´ã®å
ã¸å°ãã¾ãã"},legal:{n:"AIãªã¼ã¬ã«ã¨ã¼ã¸ã§ã³ã â Themisæè¼",d:"Themisã«çµ±æ²»ãããAIãªã¼ã¬ã«ï¼æ©å¯æ§ã»GDPRã»å®å
¨ãªç£æ»è¨¼è·¡ã®ãã¨ã§ NDAã»DPAã»MSA ã審æ»ãã¾ãã"}},eng:{ph:"ãµãããªç§ã®ãã©ãããã¯æ¨è«ã",q:"çµåãã¯ã¼ã¯ãã¼ãåãã®72éåãããã«ã¼ãã£ã³ã°ã",t:"ç±åå¦ãµã³ããªã³ã°ã精度99.7ï¼
ã",m:"䏦忍è«ã®ããã®50ä¸åæ£ã¨ã¼ã¸ã§ã³ãã"},val:{tasks:{k:"èªååãããã¿ã¹ã¯",v:"å復ã¯ã¼ã¯ããã¼ã¯èªå¾ã¨ã¼ã¸ã§ã³ãã«å§è²ã"},time:{k:"åãæ»ããæé",v:"ãã¼ã ã¯å
¥åã§ã¯ãªãæææ±ºå®ã«éä¸ã"},cost:{k:"ã³ã¹ã忏",v:"Infernex Coreã¯SLAãæºããæãå®ä¾¡ãªã¨ã³ã¸ã³ã«ã«ã¼ãã£ã³ã°ã"},roi:{k:"測å®å¯è½ãªROI",v:"ãã¹ã¦ã®è¡åã¯ãã°åã»ç£æ»å¯è½ã»äºæ¥ææã«ç´ä»ãããã¾ãã"}},s1:{eyebrow:"AIã¯ã¼ã¯ãã©ã¼ã¹ã¨ã¯",title:"AIãã¼ã«ã§ã¯ãªããAI徿¥å¡ã",body:"Infernexã®åAI徿¥å¡ã¯ãæç¢ºãªå½¹å²ã»ç¬èªã®æç¤ºã»ãã¼ã«ã¸ã®ã¢ã¯ã»ã¹ã»ãã¸ãã¹ã®é·æè¨æ¶ãæã¡ã¾ãã",f1:{t:"ããã³ããã§ã¯ãªãå½¹å²",d:"å徿¥å¡ã¯æ¥åæ©è½ã«çµãããç®æ¨ã»KPIã»ã¬ã¼ãã¬ã¼ã«ãæã¡ã¾ãã"}
811,f2:{t:"ã¹ã¿ãã¯ã«æ¥ç¶",d:"CRMã»ãã«ããã¹ã¯ã»ã¡ã¼ã«ã»ããã¥ã¡ã³ãã»DBãè¡åã¯ææ¡ã§ã¯ãªãå®è¡ããã¾ãã"},f3:{t:"統治ãããç£æ»å¯è½",d:"ãã¹ã¦ã®è¡åããã°åããã¼ã«ãã¼ã¹ã®ã¢ã¯ã»ã¹ãGDPRæºæ ãã¹ã¤ã¹è£½ã"}},s2:{eyebrow:"æ°ãæã§ã¯ãªãæ°åã§",title:"3ã¹ãããã§AI徿¥å¡ã使ã",st1:{t:"å½¹å²ãé¸ã¶",d:"å¶æ¥ã»ãµãã¼ãã»ãã¼ã±ã»è²¡åã»ãªãµã¼ã â ãããã¯ã«ã¹ã¿ã å½¹å²ã"},st2:{t:"ãã¼ã«ãæ¥ç¶",d:"CRMã»ãã«ããã¹ã¯ã»ããã¥ã¡ã³ãã»DBã¸ã®ã¯ã³ã¯ãªãã¯é£æºã"},st3:{t:"ãããã¤ãã¦å復",d:"徿¥å¡ãä»äºãéå§ãè¡åã確èªãã¬ã¼ãã¬ã¼ã«ã調æ´ãã¹ã±ã¼ã«ã"},cta:"徿¥å¡ãã«ãã¼ã試ã"},s3:{eyebrow:"å³ãããã¤å¯è½ãªå½¹å²",title:"AIãã¼ã ã¡ã¤ãã«ä¼ãã¾ãããã",explore:"å½¹å²ãè¦ã"},s4:{eyebrow:"ãªããã¼ã ã¯Infernexãé¸ã¶ã",title:"è¦ãããã®ææ¨ã§ã¯ãªããæ¸¬å®å¯è½ãªææã",body:"ä½çºçãªãã³ããã¼ã¯ã¯å
¬éãã¾ãããå®éã®ROIã¯ã¯ã¼ã¯ããã¼ã¨éã«ä¾åãã¾ãã"},s5:{eyebrow:"å
å´ã®ã¨ã³ã¸ã³",title:"Infernex Coreãé§åã",body:"åAI徿¥å¡ã¯Infernex Coreä¸ã§åä½ããåã¿ã¹ã¯ããã©ãããã¯ã»éåã»ç±åå¦ã»MiroFishã¨ã³ã¸ã³ã«æ¯ãåãã¾ãã",cta:"ãããããã¼ãã©ãããã©ã¼ã ã¸"},s6:{eyebrow:"ã¯ã¼ã¯ãã©ã¼ã¹ vs æ§æ¥ã®ããæ¹",title:"1ã¤ã®ãã©ãããã©ã¼ã ãåãé¢ããã10åã®ãã¼ã«ã§ã¯ããã¾ããã",old:{h:"徿¥ã®ã¹ã¿ãã¯",t:"夿°ã®ã½ããã»ãµã¤ãã»èããªã¼ãã¡ã¼ã·ã§ã³ã",i1:"æ©è½ãã¨ã«å¥ã
ã®SaaSã",i2:"API夿´ã§å£ããZapierãã§ã¼ã³ã",i3:"æ¥åè¨æ¶ãæããªãå¤ç«ããAIãã¼ã«ã",i4:"ã³ã¹ãã¯ç©ã¿ä¸ãããã¬ããã³ã¹ã¯å¤±ãããã"},new:{h:"Infernex AIã¯ã¼ã¯ãã©ã¼ã¹",t:"1ã¤ã®ã¯ã¼ã¯ãã©ã¼ã¹ã1ã¤ã®ãã©ãããã©ã¼ã ã1ã¤ã®è«æ±æ¸ã",i1:"æ¥åæ©è½ãã¨ã®èªå¾AI徿¥å¡ã",i2:"CRMã»ããã¥ã¡ã³ãã»ãã«ããã¹ã¯ã®çµ±åã³ãã¯ã¿ã",i3:"ãã¼ã¿ã¨ããã»ã¹ã®é·æè¨æ¶ã",i4:"ç£æ»å¯è½ãªè¡åããã¼ã«ãã¼ã¹ã®ã¬ããã³ã¹ã1ã¤ã®è«æ±æ¸ã"},cta:"æåã®AI徿¥å¡ã使",microcopy:"æ°¸ä¹
ç¡æãã©ã³ããã¤ã§ãã¢ããã°ã¬ã¼ãã"},faq:{eyebrow:"ãããã質å",title:"ç¥ã£ã¦ããã¹ããã¹ã¦",subtitle:"èªä¿¡ãæã£ã¦AIã¯ã¼ã¯ãã©ã¼ã¹ãå±éã»çµ±æ²»ã»æ¡å¼µã",q1:"AI徿¥å¡ã¨ã¯æ£ç¢ºã«ä½ã§ããï¼",a1:"AI徿¥å¡ã¯ãæç¢ºãªå½¹å²ã»æç¤ºã»ãã¼ã«æ¨©éã»æ¥åã®é·æè¨æ¶ãåããèªå¾ã¨ã¼ã¸ã§ã³ãã§ãã",q2:"æåã®AI徿¥å¡ã®å±éã«ã©ãããããããã¾ããï¼",a2:"5åæªæºããã³ãã¬ãé¸ã³ããã¼ã¿ã½ã¼ã¹ãæ¥ç¶ããããã·ã§ã³ã¨æ¨©éã確èªãã¦å±éãã¾ãã",q3:"ä¼ç¤¾ãã¼ã¿ã¯å®å
¨ã§GDPR対å¿ã§ããï¼",a3:"ã¯ããå
¨è¡åã¯ãã°åã»çµç¹åä½ã§éé¢ãããæ©å¾®ãªæä½ã¯äººææ¿èªã§ä¿è·ããã¾ãã",q4:"AI徿¥å¡ã«å¯¾ãã¦äººéã®ã³ã³ããã¼ã«ã¯ç¶æã§ãã¾ããï¼",a4:"常ã«ããã¼ã«ã»è¡åã»ãã¼ã¿ã½ã¼ã¹ãã¨ã«æ¨©éãå®ç¾©ãã¾ãã",q5:"ã©ã®ãã¼ã«/CRMã¨æ¥ç¶ã§ãã¾ããï¼",a5:"Salesforce, HubSpot, Zoho, Pipedrive, Slack, Gmail, Google Drive, Notion, DB, Stripe, Zendeskã»ã30以ä¸ã",q6:"æéä½ç³»ã¯ï¼",a6:"éå®å©ç¨ã§ç¡æéå§ãææãã©ã³ã§å¾æ¥å¡ã»ãã¼ã¯ã³ã»ã³ãã¯ã¿ãæ¡å¼µããã¾ãã",q7:"èªåã®LLMã使ã£ããç¹å®ã¢ãã«ã¸ã«ã¼ãã£ã³ã°ã§ãã¾ããï¼",a7:"å¯è½ãInfernex Coreã¯åã¿ã¹ã¯ãæãå¹ççãªã¨ã³ã¸ã³ã¾ãã¯æå®ãããã¤ãã¸ã«ã¼ãã£ã³ã°ãã¾ãã",q8:"ç¾å®çã«ã©ã®ç¨åº¦ã®ROIãæå¾
ã§ãã¾ããï¼",a8:"å¤ãã®ãã¼ã ã¯æåã®60æ¥ã§å復使¥ã40â70ï¼
忏ãã¦ãã¾ãã",cta:"ãã¼ã ã«ç¸è«"}}},hi:{nav:{features:"विशà¥à¤·à¤¤à¤¾à¤à¤",savings:"बà¤à¤¤",testimonials:"पà¥à¤°à¤¶à¤à¤¸à¤¾à¤ªà¤¤à¥à¤°",pricing:"मà¥à¤²à¥à¤¯",faq:"सामानà¥à¤¯ पà¥à¤°à¤¶à¥à¤¨",language:"à¤à¤¾à¤·à¤¾",login:"लà¥à¤à¤¿à¤¨",signup:"साà¤à¤¨ à¤
प",products:"à¤à¤¤à¥à
811¤ªà¤¾à¤¦",closeMenu:"मà¥à¤¨à¥ बà¤à¤¦ à¤à¤°à¥à¤",openMenu:"मà¥à¤¨à¥ à¤à¥à¤²à¥à¤",support:"सहायता",blog:"बà¥à¤²à¥à¤",developers:"डà¥à¤µà¤²à¤ªà¤°à¥à¤¸",workforce:"AI वरà¥à¤à¤«à¥à¤°à¥à¤¸",hire:"हायर à¤à¤°à¥à¤",workforcePricing:"मà¥à¤²à¥à¤¯",tagAutonomous:"सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ à¤à¤°à¥à¤®à¤à¤¾à¤°à¥",tagAgents12:"12 AI à¤à¥à¤®à¥à¤ªà¤²à¥à¤",tagStarter99:"â¬59/माह सà¥",tagLegalAI:"à¤à¤¾à¤¨à¥à¤¨à¥ AI",inference:"à¤à¤¨à¥à¤«à¤°à¥à¤à¤¸ API",tagOpenAI:"OpenAI-सà¤à¤à¤¤",tagBuild:"Infernex सॠबनाà¤à¤"},institutionalFooter:{badgeLabels:{gdpr:"GDPR à¤
नà¥à¤°à¥à¤ª",tls13:"TLS 1.3 à¤à¤¨à¥à¤à¥à¤°à¤¿à¤ªà¥à¤¶à¤¨",swissHosted:"सà¥à¤µà¤¿à¤à¥à¤à¤°à¤²à¥à¤à¤¡-हà¥à¤¸à¥à¤à¥à¤¡ à¤à¤¨à¥à¤«à¥à¤°à¤¾à¤¸à¥à¤à¥à¤°à¤à¥à¤à¤°",rlsAuth:"पà¤à¤à¥à¤¤à¤¿-सà¥à¤¤à¤° सà¥à¤°à¤à¥à¤·à¤¾"}},wh:{seo:{title:"Infernex AI â à¤
पनॠसà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI वरà¥à¤à¤«à¥à¤°à¥à¤¸ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¥à¤",desc:"Infernex AI Workforce à¤à¤ªà¤à¥ डà¥à¤à¤¾, à¤à¥à¤² à¤à¤° पà¥à¤°à¤à¥à¤°à¤¿à¤¯à¤¾à¤à¤ à¤à¥ à¤à¥à¤¡à¤¼à¤à¤° 24/7 à¤à¤ªà¤à¥ à¤à¥à¤® à¤à¥ साथ à¤à¤¾à¤® à¤à¤°à¤¨à¥ वालॠAI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¤¤à¤¾ हà¥à¥¤"},hero:{badge:"Infernex AI वरà¥à¤à¤«à¥à¤°à¥à¤¸",line1:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ नियà¥à¤à¥à¤¤ à¤à¤°à¥à¤ à¤à¤° à¤
पनॠà¤à¤à¤ªà¤¨à¥ बनाà¤à¤",line1Accent:"à¤à¤ à¤à¥à¤²à¤¿à¤ मà¥à¤",line2:"à¤à¤ विà¤à¤¾à¤° सॠà¤à¤µà¤¿à¤·à¥à¤¯ à¤à¥ Fortune 500 à¤à¤à¤ªà¤¨à¥ तà¤à¥¤",moreInfo:"à¤à¤° à¤à¤¾à¤¨à¥à¤",ctaPrimary:"मà¥à¤°à¤¾ पहला AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ बनाà¤à¤ â निà¤à¤¶à¥à¤²à¥à¤",ctaSecondary:"Infernex API सॠबनाà¤à¤",microcopy:"à¤à¥à¤°à¥à¤¡à¤¿à¤ à¤à¤¾à¤°à¥à¤¡ नहà¥à¤à¥¤ 50K मà¥à¤«à¥à¤¤ à¤à¥à¤à¤¨/माह। OpenAI SDK सà¤à¤à¤¤à¥¤",valueProp:"à¤à¤à¤ªà¤¨à¤¿à¤¯à¤¾à¤ हर साल, दà¥à¤¹à¤°à¤¾à¤ à¤à¤¾à¤¨à¥ वालॠà¤à¤¾à¤® मà¥à¤ लाà¤à¥à¤ डà¥à¤²à¤° à¤à¥ दà¥à¤¤à¥ हà¥à¤à¥¤ Infernex सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ à¤à¥ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¤¤à¤¾ हॠà¤à¥ à¤à¤ªà¤à¥ वà¥à¤¯à¤µà¤¸à¤¾à¤¯ à¤à¥ 24/7 सà¥à¤µà¤à¤¾à¤²à¤¿à¤¤ à¤à¤°à¤¤à¥ हà¥à¤à¥¤"},emp:{sales:{n:"AI बिà¤à¥à¤°à¥ à¤à¤à¥à¤à¤",d:"à¤à¤¨à¤¬à¤¾à¤à¤à¤¡ लà¥à¤¡à¥à¤¸ à¤à¥ à¤à¥à¤µà¤¾à¤²à¤¿à¤«à¤¾à¤ à¤à¤°à¤¤à¤¾ हà¥, à¤à¤à¤à¤°à¥à¤ à¤à¥ परà¥à¤¸à¤¨à¤²à¤¾à¤à¤à¤¼ à¤à¤°à¤¤à¤¾ हॠà¤à¤° à¤à¤ªà¤à¤¾ CRM रà¥à¤¯à¤²-à¤à¤¾à¤à¤® मà¥à¤ à¤
पडà¥à¤ à¤à¤°à¤¤à¤¾ हà¥à¥¤"},support:{n:"AI à¤à¥à¤°à¤¾à¤¹à¤ सहायता",d:"L1 à¤à¤¿à¤à¤ 24/7 सà¥à¤²à¤à¤¾à¤¤à¤¾ हà¥, सà¤à¤µà¥à¤¦à¤¨à¤¶à¥à¤² मामलà¥à¤ à¤à¥ à¤à¤¸à¥à¤à¥à¤²à¥à¤ à¤à¤°à¤¤à¤¾ हॠà¤à¤° à¤à¤ªà¤à¥ हà¥à¤²à¥à¤ªà¤¡à¥à¤¸à¥à¤ सॠà¤à¥à¤¡à¤¼à¤¤à¤¾ हà¥à¥¤"},marketing:{n:"AI मारà¥à¤à¥à¤à¤¿à¤à¤ à¤à¤à¥à¤à¤",d:"à¤
à¤à¤¿à¤¯à¤¾à¤¨ यà¥à¤à¤¨à¤¾ बनाता हà¥, मलà¥à¤à¥-à¤à¥à¤¨à¤² à¤à¥à¤ªà¥ लिà¤à¤¤à¤¾ हà¥, पà¥à¤°à¤¦à¤°à¥à¤¶à¤¨ à¤à¥à¤°à¥à¤ à¤à¤°à¤¤à¤¾ हॠà¤à¤° सà¥à¤§à¤¾à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤"},finance:{n:"AI वितà¥à¤¤ à¤à¤à¥à¤à¤",d:"à¤à¤¨à¥à¤µà¥à¤à¤¸ मिलान, à¤à¥à¤¶à¤«à¥à¤²à¥ पà¥à¤°à¥à¤µà¤¾à¤¨à¥à¤®à¤¾à¤¨ à¤à¤° लाà¤à¤¤ बननॠसॠपहलॠà¤
सामानà¥à¤¯à¤¤à¤¾à¤à¤ à¤à¥ सà¥à¤à¤¨à¤¾à¥¤"},research:{n:"AI शà¥à¤§ à¤à¤à¥à¤à¤",d:"सà¥à¤°à¥à¤¤à¥à¤ à¤à¥ सà¥à¤à¥à¤¨ à¤à¤°à¤¤à¤¾ हà¥, बà¥à¤°à¥à¤« बनाता हॠà¤à¤° à¤à¤ªà¤à¥ à¤à¥à¤® à¤à¥ बाà¤à¤¾à¤° सॠà¤à¤à¥ रà¤à¤¤à¤¾ हà¥à¥¤"},legal:{n:"AI à¤à¤¾à¤¨à¥à¤¨à¥ à¤à¤à¥à¤à¤ â Themis सà¤à¤à¤¾à¤²à¤¿à¤¤",d:"Themis-शासित AI à¤à¤¾à¤¨à¥à¤¨à¥ सहायà¤: à¤à¥à¤ªà¤¨à¥à¤¯à¤¤à¤¾, GDPR à¤à¤° पà¥à¤°à¥à¤£ à¤à¤¡à¤¿à¤ à¤à¥à¤°à¥à¤² à¤à¥ तहत NDA, DPA à¤à¤° MSA à¤à¥ समà¥à¤à¥à¤·à¤¾ à¤à¤°à¤¤à¤¾ हà¥à¥¤"}},eng:{ph:"सब-मिलà¥à¤¸à¥à¤à¤à¤¡ लà¥à¤à¥à¤à¤¸à¥ à¤à¥ लिठफà¥à¤à¥à¤¨à¤¿à¤ à¤à¤¨à¥à¤«à¤°à¥à¤à¤¸à¥¤",q:"à¤à¥à¤®à¥à¤¬à¤¿à¤¨à¥à¤à¥à¤°à¤¿à¤¯à¤² वरà¥à¤à¤²à¥à¤¡ à¤à¥ लिठ72-à¤à¥à¤¯à¥à¤¬à¤¿à¤ à¤à¥à¤µà¤¾à¤à¤à¤® रà¥à¤à¤¿à¤à¤à¥¤",t:"थरà¥à¤®à¥à¤¡à¤¾à¤¯à¤¨à¤¾à¤®à¤¿à¤ सà¥à¤à¤ªà¤²à¤¿à¤à¤, 99.7% परिशà¥à¤¦à¥à¤§à¤¤à¤¾à¥¤",m:"समानाà¤à¤¤à¤° तरà¥à¤ à¤à¥ लिठ5,00,000 वितरित à¤à¤à¥à¤à¤à¥¤"},val:{tasks:{k:"सà¥à¤µà¤à¤¾à¤²à¤¿à¤¤ à¤à¤¾à¤°à¥à¤¯",v:"दà¥à¤¹à¤°à¤¾à¤µ वालॠपà¥à¤°à¤µà¤¾à¤¹ सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ à¤à¤à¥à¤à¤à¥à¤ à¤à¥ सà¥à¤à¤ªà¥ à¤à¤à¥¤"}
811,time:{k:"पà¥à¤¨à¤ पà¥à¤°à¤¾à¤ªà¥à¤¤ समय",v:"à¤à¤ªà¤à¥ à¤à¥à¤® डà¥à¤à¤¾ à¤à¤à¤à¥à¤°à¥ नहà¥à¤, निरà¥à¤£à¤¯ पर à¤à¥à¤à¤¦à¥à¤°à¤¿à¤¤à¥¤"},cost:{k:"लाà¤à¤¤ मà¥à¤ à¤à¤®à¥",v:"Infernex Core à¤à¤ªà¤à¥ SLA à¤à¥ पà¥à¤°à¤¾ à¤à¤°à¤¨à¥ वालॠसबसॠससà¥à¤¤à¥ à¤à¤à¤à¤¨ à¤à¥ रà¥à¤ à¤à¤°à¤¤à¤¾ हà¥à¥¤"},roi:{k:"मापन यà¥à¤à¥à¤¯ ROI",v:"हर à¤à¤¾à¤°à¥à¤¯ लà¥à¤, à¤à¤¡à¤¿à¤ यà¥à¤à¥à¤¯ à¤à¤° वà¥à¤¯à¤¾à¤µà¤¸à¤¾à¤¯à¤¿à¤ परिणाम सॠà¤à¥à¤¡à¤¼à¤¾à¥¤"}},s1:{eyebrow:"AI वरà¥à¤à¤«à¥à¤°à¥à¤¸ à¤à¥à¤¯à¤¾ हà¥",title:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥, à¤à¥à¤µà¤² AI à¤à¥à¤² नहà¥à¤à¥¤",body:"हर Infernex AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¥ à¤à¤ निरà¥à¤§à¤¾à¤°à¤¿à¤¤ à¤à¥à¤®à¤¿à¤à¤¾, à¤
पनॠनिरà¥à¤¦à¥à¤¶, à¤à¤ªà¤à¥ à¤à¥à¤² तठपहà¥à¤à¤ à¤à¤° à¤à¤ªà¤à¥ वà¥à¤¯à¤µà¤¸à¤¾à¤¯ à¤à¥ दà¥à¤°à¥à¤à¤à¤¾à¤²à¤¿à¤ सà¥à¤®à¥à¤¤à¤¿ हà¥à¤¤à¥ हà¥à¥¤",f1:{t:"à¤à¤ à¤à¥à¤®à¤¿à¤à¤¾, पà¥à¤°à¥à¤®à¥à¤ªà¥à¤ नहà¥à¤",d:"हर à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¤ वà¥à¤¯à¤¾à¤µà¤¸à¤¾à¤¯à¤¿à¤ à¤à¤¾à¤°à¥à¤¯ पर à¤à¥à¤à¤¦à¥à¤°à¤¿à¤¤ हॠâ लà¤à¥à¤·à¥à¤¯, KPI à¤à¤° सà¥à¤®à¤¾à¤à¤ à¤à¥ साथ।"},f2:{t:"à¤à¤ªà¤à¥ सà¥à¤à¥à¤ सॠà¤à¥à¤¡à¤¼à¤¾",d:"CRM, हà¥à¤²à¥à¤ªà¤¡à¥à¤¸à¥à¤, à¤à¤®à¥à¤², डà¥à¤à¥à¤¸, डà¥à¤à¤¾à¤¬à¥à¤¸à¥¤ à¤à¤¾à¤°à¥à¤¯ à¤à¥à¤µà¤² सà¥à¤à¤¾à¤ नहà¥à¤, निषà¥à¤ªà¤¾à¤¦à¤¿à¤¤ हà¥à¤¤à¥ हà¥à¤à¥¤"},f3:{t:"शासित à¤à¤° à¤à¤¡à¤¿à¤ यà¥à¤à¥à¤¯",d:"हर à¤à¤¾à¤°à¥à¤¯ लà¥à¤à¥¤ à¤à¥à¤®à¤¿à¤à¤¾ à¤à¤§à¤¾à¤°à¤¿à¤¤ पहà¥à¤à¤à¥¤ GDPR à¤
नà¥à¤°à¥à¤ªà¥¤ सà¥à¤µà¤¿à¤¸ à¤à¤à¤à¥à¤¨à¤¿à¤¯à¤°à¤¿à¤à¤à¥¤"}},s2:{eyebrow:"मिनà¤à¥à¤ मà¥à¤, महà¥à¤¨à¥à¤ मà¥à¤ नहà¥à¤",title:"तà¥à¤¨ à¤à¤°à¤£à¥à¤ मà¥à¤ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ बनाà¤à¤à¥¤",st1:{t:"à¤à¥à¤®à¤¿à¤à¤¾ à¤à¥à¤¨à¥à¤",d:"बिà¤à¥à¤°à¥, सहायता, मारà¥à¤à¥à¤à¤¿à¤à¤, वितà¥à¤¤, शà¥à¤§ â या à¤à¤¸à¥à¤à¤® à¤à¥à¤®à¤¿à¤à¤¾à¥¤"},st2:{t:"à¤
पनॠà¤à¥à¤² à¤à¥à¤¡à¤¼à¥à¤",d:"CRM, हà¥à¤²à¥à¤ªà¤¡à¥à¤¸à¥à¤, डà¥à¤à¥à¤¸ à¤à¤° DB à¤à¥ लिठà¤à¤-à¤à¥à¤²à¤¿à¤ à¤à¤¨à¥à¤à¥à¤à¤°à¥¤"},st3:{t:"तà¥à¤¨à¤¾à¤¤ à¤à¤°à¥à¤ à¤à¤° सà¥à¤§à¤¾à¤°à¥à¤",d:"à¤à¤ªà¤à¤¾ à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¤¾à¤® शà¥à¤°à¥ à¤à¤°à¤¤à¤¾ हà¥à¥¤ à¤à¤¾à¤°à¥à¤¯ समà¥à¤à¥à¤·à¤¾ à¤à¤°à¥à¤, सà¥à¤®à¤¾à¤à¤ समायà¥à¤à¤¿à¤¤ à¤à¤°à¥à¤, सà¥à¤à¥à¤² à¤à¤°à¥à¤à¥¤"},cta:"à¤à¤®à¥à¤ªà¥à¤²à¥à¤ बिलà¥à¤¡à¤° à¤à¤à¤¼à¤®à¤¾à¤à¤"},s3:{eyebrow:"तà¥à¤°à¤à¤¤ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¤¨à¥ यà¥à¤à¥à¤¯ à¤à¥à¤®à¤¿à¤à¤¾à¤à¤",title:"à¤
पनॠAI साथियà¥à¤ सॠमिलà¥à¤à¥¤",explore:"à¤à¥à¤®à¤¿à¤à¤¾ दà¥à¤à¥à¤"},s4:{eyebrow:"à¤à¥à¤®à¥à¤ Infernex à¤à¥à¤¯à¥à¤ à¤à¥à¤¨à¤¤à¥ हà¥à¤",title:"मापन यà¥à¤à¥à¤¯ परिणाम, दिà¤à¤¾à¤µà¤à¥ मà¥à¤à¥à¤°à¤¿à¤à¥à¤¸ नहà¥à¤à¥¤",body:"हम मनà¤à¤¢à¤¼à¤à¤¤ बà¥à¤à¤à¤®à¤¾à¤°à¥à¤ पà¥à¤°à¤à¤¾à¤¶à¤¿à¤¤ नहà¥à¤ à¤à¤°à¤¤à¥à¥¤ वासà¥à¤¤à¤µà¤¿à¤ ROI à¤à¤ªà¤à¥ पà¥à¤°à¤à¥à¤°à¤¿à¤¯à¤¾à¤à¤ à¤à¤° वà¥à¤²à¥à¤¯à¥à¤® पर निरà¥à¤à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤"},s5:{eyebrow:"à¤
à¤à¤¦à¤° à¤à¤¾ à¤à¤à¤à¤¨",title:"Infernex Core दà¥à¤µà¤¾à¤°à¤¾ सà¤à¤à¤¾à¤²à¤¿à¤¤à¥¤",body:"हर AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ Infernex Core पर à¤à¤²à¤¤à¤¾ हॠâ à¤à¤ बà¥à¤¦à¥à¤§à¤¿à¤®à¤¾à¤¨ राà¤à¤à¤° à¤à¥ हर à¤à¤¾à¤°à¥à¤¯ à¤à¥ फà¥à¤à¥à¤¨à¤¿à¤, à¤à¥à¤µà¤¾à¤à¤à¤®, थरà¥à¤®à¥à¤¡à¤¾à¤¯à¤¨à¤¾à¤®à¤¿à¤ या MiroFish à¤à¤à¤à¤¨ à¤à¥ à¤à¥à¤à¤¤à¤¾ हà¥à¥¤",cta:"डà¥à¤µà¤²à¤ªà¤° पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤® दà¥à¤à¥à¤"},s6:{eyebrow:"वरà¥à¤à¤«à¥à¤°à¥à¤¸ बनाम पà¥à¤°à¤¾à¤¨à¤¾ तरà¥à¤à¤¾",title:"à¤à¤ पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤®à¥¤ दस à¤
सà¤à¤¬à¤¦à¥à¤§ à¤à¥à¤² नहà¥à¤à¥¤",old:{h:"à¤à¥à¤²à¤¾à¤¸à¤¿à¤ सà¥à¤à¥à¤",t:"à¤à¤ सà¥à¤«à¥à¤à¤µà¥à¤¯à¤°, साà¤à¤²à¥, नाà¤à¤¼à¥à¤ à¤à¤à¥à¤®à¥à¤¶à¤¨à¥¤",i1:"हर à¤à¤¾à¤°à¥à¤¯ à¤à¥ लिठà¤
लठSaaS।",i2:"API बदलनॠपर à¤à¥à¤à¤¨à¥ वालॠZapier à¤à¥à¤¨à¥¤",i3:"वà¥à¤¯à¤µà¤¸à¤¾à¤¯ à¤à¥ सà¥à¤®à¥à¤¤à¤¿ à¤à¥ बिना à¤
लà¤-थलठAI à¤à¥à¤²à¥¤",i4:"लाà¤à¤¤ बढ़तॠहà¥, शासन à¤à¥à¤¤à¤¾ हà¥à¥¤"},new:{h:"Infernex AI वरà¥à¤à¤«à¥à¤°à¥à¤¸",t:"à¤à¤ वरà¥à¤à¤«à¥à¤°à¥à¤¸, à¤à¤ पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤®, à¤à¤ बिल।",i1:"हर वà¥à¤¯à¤¾à¤µà¤¸à¤¾à¤¯à¤¿à¤ à¤à¤¾à¤°à¥à¤¯ à¤à¥ लिठसà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥à¥¤",i2:"CRM, डà¥à¤à¥à¤¸ à¤à¤° हà¥à¤²à¥à¤ªà¤¡à¥à¤¸à¥à¤ à¤à¥ लिठà¤à¤à¥à¤à¥à¤¤ à¤à¤¨à¥à¤à¥à¤à¤°à¥¤",i3:"à¤à¤ªà¤à¥ डà¥à¤à¤¾ à¤à¤° पà¥à¤°à¤à¥à¤°à¤¿à¤¯à¤¾à¤à¤ à¤à¥ दà¥à¤°à¥à¤à¤à¤¾à¤²à¤¿à¤ सà¥à¤®à¥à¤¤à¤¿à¥¤",i4:"à¤à¤¡à¤¿à¤ यà¥à¤à¥à¤¯ à¤à¤¾à¤°à¥à¤¯, à¤à¥à¤®à¤¿à¤à¤¾ à¤à¤§à¤¾à¤°à¤¿à¤¤ शासन, à¤à¤ बिल।"},cta:"मà¥à¤°à¤¾ पहला AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ बनाà¤à¤",microcopy:"हमà¥à¤¶à¤¾ à¤à¥ लिठमà¥à¤«à¥à¤¤ पà¥à¤²à¤¾à¤¨à¥¤ à¤à¤à¥ à¤à¥ à¤
पà¤à¥à¤°à¥à¤¡ à¤à¤°à¥à¤à¥¤"},faq:{eyebrow:"à¤
à¤à¥à¤¸à¤° पà¥à¤à¥ à¤à¤¾à¤¨à¥ वालॠपà¥à¤°à¤¶à¥à¤¨",title:"à¤à¤ªà¤à¥ à¤à¥ à¤à¤¾à¤¨à¤¨à¤¾ à¤à¤¼à¤°à¥à¤°à¥ हà¥",subtitle:"à¤à¤¤à¥à¤®à¤µà¤¿à¤¶à¥à¤µà¤¾à¤¸ à¤à¥ साथ à¤
पनॠAI वरà¥à¤à¤«à¥à¤°à¥à¤¸ तà¥à¤¨à¤¾à¤¤, शासित à¤à¤° सà¥à¤à¥à¤² à¤à¤°à¥à¤à¥¤",q1:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ वासà¥à¤¤à¤µ मà¥à¤ à¤à¥à¤¯à¤¾ हà¥?",a1:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¤ सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ à¤à¤à¥à¤à¤ हॠà¤à¤¿à¤¸à¤à¥ निरà¥à¤§à¤¾à¤°à¤¿à¤¤ à¤à¥à¤®à¤¿à¤à¤¾, निरà¥à¤¦à¥à¤¶, à¤à¥à¤² à¤
नà¥à¤®à¤¤à¤¿à¤¯à¤¾à¤ à¤à¤° à¤à¤ªà¤à¥ वà¥à¤¯à¤µà¤¸à¤¾à¤¯ à¤à¥ दà¥à¤°à¥à¤à¤à¤¾à¤²à¤¿à¤ सà¥à¤®à¥à¤¤à¤¿ हà¥à¤¤à¥ हà¥à¥¤",q2:"पहला AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¤¨à¥ मà¥à¤ à¤à¤¿à¤¤à¤¨à¤¾ समय लà¤à¤¤à¤¾ हà¥?",a2:"5 मिनठसॠà¤à¤®à¥¤ à¤à¥à¤®à¥à¤ªà¤²à¥à¤ à¤à¥à¤¨à¥à¤, डà¥à¤à¤¾ सà¥à¤°à¥à¤¤ à¤à¥à¤¡à¤¼à¥à¤, मिशन à¤à¤° à¤
नà¥à¤®à¤¤à¤¿à¤¯à¤¾à¤ à¤à¤¾à¤à¤à¥à¤, तà¥à¤¨à¤¾à¤¤ à¤à¤°à¥à¤à¥¤",q3:"à¤à¥à¤¯à¤¾ मà¥à¤°à¤¾ डà¥à¤à¤¾ सà¥à¤°à¤à¥à¤·à¤¿à¤¤ à¤à¤° GDPR-तà¥à¤¯à¤¾à¤° हà¥?",a3:"हाà¤à¥¤ हर à¤à¤¾à¤°à¥à¤¯ लà¥à¤ हà¥à¤¤à¤¾ हà¥, सà¤à¤à¤ न à¤à¥ à¤
नà¥à¤¸à¤¾à¤° पà¥à¤¥à¤ हॠà¤à¤° सà¤à¤µà¥à¤¦à¤¨à¤¶à¥à¤² à¤à¤ªà¤°à¥à¤¶à¤¨à¥à¤ à¤à¥ लिठमानव सà¥à¤µà¥à¤à¥à¤¤à¤¿ दà¥à¤µà¤¾à¤°à¤¾ सà¤à¤°à¤à¥à¤·à¤¿à¤¤ हà¥à¥¤",q4:"à¤à¥à¤¯à¤¾ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ पर मानव नियà¤à¤¤à¥à¤°à¤£ बना रहता हà¥?",a4:"हमà¥à¤¶à¤¾à¥¤ à¤à¤ª हर à¤à¥à¤², à¤à¤¾à¤°à¥à¤¯ à¤à¤° डà¥à¤à¤¾ सà¥à¤°à¥à¤¤ à¤à¥ लिठà¤
नà¥à¤®à¤¤à¤¿à¤¯à¤¾à¤ परिà¤à¤¾à¤·à¤¿à¤¤ à¤à¤°à¤¤à¥ हà¥à¤à¥¤",q5:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¤¿à¤¨ à¤à¥à¤² à¤à¤° CRM सॠà¤à¥à¤¡à¤¼ सà¤à¤¤à¥ हà¥à¤?",a5:"Salesforce, HubSpot, Zoho, Pipedrive, Slack, Gmail, Google Drive, Notion, DB, Stripe, Zendesk à¤à¤° 30+ à¤
नà¥à¤¯à¥¤",q6:"मà¥à¤²à¥à¤¯ निरà¥à¤§à¤¾à¤°à¤£ à¤à¥à¤¸à¥ à¤à¤¾à¤® à¤à¤°à¤¤à¤¾ हà¥?",a6:"सà¥à¤®à¤¿à¤¤ à¤à¤ªà¤¯à¥à¤ à¤à¥ साथ निà¤à¤¶à¥à¤²à¥à¤ शà¥à¤°à¥ à¤à¤°à¥à¤à¥¤ à¤à¥à¤à¤¤à¤¾à¤¨ यà¥à¤à¤¨à¤¾à¤à¤ à¤
धिठà¤à¤°à¥à¤®à¤à¤¾à¤°à¥, à¤à¥à¤à¤¨ à¤à¤° à¤à¤¨à¥à¤à¥à¤à¤° दà¥à¤¤à¥ हà¥à¤à¥¤",q7:"à¤à¥à¤¯à¤¾ मà¥à¤ à¤
पना LLM à¤à¤ªà¤¯à¥à¤ à¤à¤° सà¤à¤¤à¤¾ हà¥à¤ या à¤à¤¿à¤¸à¥ मà¥à¤¡à¤² पर रà¥à¤ à¤à¤° सà¤à¤¤à¤¾ हà¥à¤?",a7:"हाà¤à¥¤ Infernex Core हर à¤à¤¾à¤°à¥à¤¯ à¤à¥ सबसॠà¤à¥à¤¶à¤² à¤à¤à¤à¤¨ या à¤à¤ªà¤à¥ à¤à¥à¤¨à¥ पà¥à¤°à¤¦à¤¾à¤¤à¤¾ पर रà¥à¤ à¤à¤°à¤¤à¤¾ हà¥à¥¤",q8:"मà¥à¤ वासà¥à¤¤à¤µà¤¿à¤ रà¥à¤ª सॠà¤à¥à¤¯à¤¾ ROI à¤
पà¥à¤à¥à¤·à¤¿à¤¤ à¤à¤° सà¤à¤¤à¤¾ हà¥à¤?",a8:"à¤à¥à¤®à¥à¤ à¤à¤®à¤¤à¥à¤° पर पहलॠ60 दिनà¥à¤ मà¥à¤ दà¥à¤¹à¤°à¤¾à¤µ वालॠà¤à¤¾à¤°à¥à¤¯à¤à¤¾à¤° मà¥à¤ 40â70% à¤à¥ à¤à¤®à¥ दरà¥à¤ à¤à¤°à¤¤à¥ हà¥à¤à¥¤",cta:"हमारॠà¤à¥à¤® सॠबात à¤à¤°à¥à¤"}}},ar:{nav:{features:"اÙÙ
ÙØ²Ø§Øª",savings:"Ø§ÙØªÙÙÙØ±",testimonials:"Ø§ÙØ´Ùادات",pricing:"Ø§ÙØ£Ø³Ø¹Ø§Ø±",faq:"Ø§ÙØ£Ø³Ø¦ÙØ© Ø§ÙØ´Ø§Ø¦Ø¹Ø©",language:"اÙÙØºØ©",login:"تسجÙÙ Ø§ÙØ¯Ø®ÙÙ",
811signup:"Ø¥ÙØ´Ø§Ø¡ ØØ³Ø§Ø¨",products:"اÙÙ
ÙØªØ¬Ø§Øª",closeMenu:"Ø¥ØºÙØ§Ù اÙÙØ§Ø¦Ù
Ø©",openMenu:"ÙØªØ اÙÙØ§Ø¦Ù
Ø©",support:"Ø§ÙØ¯Ø¹Ù
",blog:"اÙÙ
دÙÙØ©",developers:"اÙÙ
Ø·ÙØ±ÙÙ",workforce:"ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",hire:"ØªÙØ¸ÙÙ",workforcePricing:"Ø§ÙØ£Ø³Ø¹Ø§Ø±",tagAutonomous:"Ù
ÙØ¸ÙÙ٠ذاتÙÙÙ",tagAgents12:"12 ÙØ§Ùب Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù",tagStarter99:"Ù
Ù 59â¬/Ø´ÙØ±",tagLegalAI:"Ø°ÙØ§Ø¡ ÙØ§ÙÙÙÙ",inference:"ÙØ§Ø¬ÙØ© Ø§Ø³ØªØ¯ÙØ§Ù API",tagOpenAI:"Ù
ØªÙØ§ÙÙ Ù
ع OpenAI",tagBuild:"ابÙ٠باستخداÙ
Infernex"},institutionalFooter:{badgeLabels:{gdpr:"Ù
ØªÙØ§ÙÙ Ù
ع GDPR",tls13:"تشÙÙØ± TLS 1.3",swissHosted:"بÙÙØ© Ù
Ø³ØªØ¶Ø§ÙØ© Ù٠سÙÙØ³Ø±Ø§",rlsAuth:"Ø£Ù
ا٠عÙÙ Ù
ستÙÙ Ø§ÙØµÙ"}},wh:{seo:{title:"Infernex AI â Ø§Ùشر ÙÙØ© عÙ
ÙÙ Ø§ÙØ°Ø§ØªÙØ© Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",desc:"تربط Infernex AI Workforce Ø¨ÙØ§ÙØ§ØªÙ ÙØ£Ø¯ÙØ§ØªÙ ÙØ¹Ù
ÙÙØ§ØªÙ ÙÙØ´Ø± Ù
ÙØ¸ÙÙ Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù ÙØ¹Ù
ÙÙÙ 24/7 Ù
ع ÙØ±ÙÙ."},hero:{badge:"ÙÙØ© عÙ
Ù Infernex Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",line1:"ÙØ¸ÙÙ Ù
ÙØ¸ÙÙÙ Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù ÙØ£Ùشئ Ø´Ø±ÙØªÙ",line1Accent:"بÙÙØ±Ø© ÙØ§ØØ¯Ø©",line2:"Ù
Ù ÙÙØ±Ø© Ø¥ÙÙ Ø´Ø±ÙØ© Ù
Ø³ØªÙØ¨ÙÙØ© ضÙ
Ù Fortune 500.",moreInfo:"Ù
Ø²ÙØ¯ Ù
٠اÙÙ
عÙÙÙ
ات",ctaPrimary:"Ø£ÙØ´Ø¦ Ø£ÙÙ Ù
ÙØ¸Ù Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù â Ù
جاÙÙØ§",ctaSecondary:"ابÙ٠باستخداÙ
API Infernex",microcopy:"بدÙÙ Ø¨Ø·Ø§ÙØ© ائتÙ
اÙ. 50K تÙÙÙ Ù
جاÙÙ/Ø´ÙØ±. Ù
ØªÙØ§ÙÙ Ù
ع OpenAI SDK.",valueProp:"تخسر Ø§ÙØ´Ø±Ùات Ù
ÙØ§ÙÙÙ Ø§ÙØ¯ÙÙØ§Ø±Ø§Øª سÙÙÙÙØ§ بسبب Ø§ÙØ¹Ù
٠اÙÙ
ØªÙØ±Ø±. Infernex ÙÙØ´Ø± Ù
ÙØ¸ÙÙÙ Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù Ù
ستÙÙÙÙ ÙØ£ØªÙ
تة أعÙ
اÙÙ 24/7."},emp:{sales:{n:"ÙÙÙÙ Ù
Ø¨ÙØ¹Ø§Øª AI",d:"ÙØ¤ÙÙ Ø§ÙØ¹Ù
ÙØ§Ø¡ اÙÙ
ØØªÙ
ÙÙÙØ ÙÙØ´Ø®ØµÙ Ø§ÙØªÙØ§ØµÙØ ÙÙØØ¯ÙØ« CRM Ù٠اÙÙÙØª اÙÙØ¹ÙÙ."},support:{n:"ÙÙÙ٠دعÙ
Ø§ÙØ¹Ù
ÙØ§Ø¡ AI",d:"ÙØÙ ØªØ°Ø§ÙØ± L1 عÙÙ Ù
دار Ø§ÙØ³Ø§Ø¹Ø©Ø ÙÙØµØ¹Ùد Ø§ÙØØ§ÙØ§Øª Ø§ÙØØ³Ø§Ø³Ø© ÙÙØªØµÙ بÙ
ÙØªØ¨ اÙÙ
ساعدة."},marketing:{n:"ÙÙÙ٠تسÙÙÙ AI",d:"ÙØ®Ø·Ø· Ø§ÙØÙ
ÙØ§ØªØ ÙÙÙØªØ¨ Ù
ØØªÙÙ Ù
تعدد اÙÙÙÙØ§ØªØ ÙÙÙÙØ³ Ø§ÙØ£Ø¯Ø§Ø¡ ÙÙØªØØ³ÙÙ."},finance:{n:"ÙÙÙÙ Ù
اÙÙ AI",d:"ÙØ·Ø§Ø¨Ù اÙÙÙØ§ØªÙØ±Ø ÙÙØªÙبأ Ø¨Ø§ÙØªØ¯Ù٠اÙÙÙØ¯Ù ÙÙÙØªØ´Ù Ø§ÙØ´Ø°Ùذ ÙØ¨Ù Ø£Ù ÙÙÙÙÙÙ."},research:{n:"ÙÙÙÙ Ø¨ØØ« AI",d:"ÙÙØØµ اÙÙ
ØµØ§Ø¯Ø±Ø ÙÙÙØ®Øµ اÙÙ
ÙØ®ØµØ§Øª ÙÙÙØ¨ÙÙ ÙØ±ÙÙÙ Ù
ØªÙØ¯Ù
ÙØ§ عÙÙ Ø§ÙØ³ÙÙ."},legal:{n:"ÙÙÙÙ ÙØ§ÙÙÙÙ AI â Ù
دعÙÙ
ب٠Themis",d:"Ù
ساعد ÙØ§ÙÙÙÙ AI Ù
ØÙÙÙ
ب٠Themis: ÙØ±Ø§Ø¬Ø¹ NDA ÙDPA ÙMSA ضÙ
Ù Ø§ÙØ³Ø±ÙØ© ÙGDPR ÙÙ
سار تدÙÙÙ ÙØ§Ù
Ù."}},eng:{ph:"Ø§Ø³ØªØ¯ÙØ§Ù ÙÙØªÙÙ٠بزÙ
٠استجابة دÙ٠اÙÙ
ÙÙ٠ثاÙÙØ©.",q:"ØªÙØ¬ÙÙ ÙÙ
ÙÙ
٠ب٠72 ÙÙÙØ¨Øª ÙÙØ£ØÙ
Ø§Ù Ø§ÙØªÙاÙÙÙØ©.",t:"أخذ عÙÙØ§Øª ثرÙ
ÙØ¯ÙÙØ§Ù
ÙÙÙ Ø¨Ø¯ÙØ© 99.7Ùª.",m:"500,000 ÙÙÙÙ Ù
ÙØ²Ø¹ ÙÙØ§Ø³ØªØ¯Ùا٠اÙÙ
ØªÙØ§Ø²Ù."},val:{tasks:{k:"اÙÙ
ÙØ§Ù
اÙÙ
ؤتÙ
تة",v:"تسÙÙÙ
Ø³ÙØ± Ø§ÙØ¹Ù
٠اÙÙ
ØªÙØ±Ø± Ø¥ÙÙ ÙÙÙØ§Ø¡ ذاتÙÙÙ."},time:{k:"اÙÙÙØª اÙÙ
ستعاد",v:"ÙØ±Ùز ÙØ±ÙÙ٠عÙ٠اÙÙØ±Ø§Ø±Ø§ØªØ ÙØ§ عÙÙ Ø¥Ø¯Ø®Ø§Ù Ø§ÙØ¨ÙØ§ÙØ§Øª."},cost:{k:"Ø®ÙØ¶ Ø§ÙØªÙاÙÙÙ",v:"ÙÙØ¬ÙÙ Infernex Core Ø¥Ù٠أرخص Ù
ØØ±Ù ÙÙØ¨Ù SLA."},roi:{k:"عائد ÙØ§Ø¨Ù ÙÙÙÙØ§Ø³",v:"ÙÙØ³Ø¬ÙÙÙ Ù٠إجراء ÙÙÙ
Ù٠تدÙÙÙÙ ÙÙ
رتبط Ø¨ÙØªÙجة عÙ
Ù."}},s1:{eyebrow:"Ù
ا ÙÙ ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",title:"Ù
ÙØ¸ÙÙ Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹ÙØ ÙÙÙØ³ Ù
جرد Ø£Ø¯ÙØ§Øª.",body:"ÙÙÙ Ù
ÙØ¸Ù AI ÙÙ Infernex Ø¯ÙØ± Ù
ØØ¯Ø¯ ÙØªØ¹ÙÙÙ
ات خاصة ÙØµÙاØÙات عÙÙ Ø£Ø¯ÙØ§ØªÙ ÙØ°Ø§Ùرة Ø·ÙÙÙØ© Ø§ÙØ£Ù
د ÙØ£Ø¹Ù
اÙÙ.",f1:{t:"Ø¯ÙØ±Ø ÙÙÙØ³ برÙÙ
بت",d:"ÙÙ Ù
ÙØ¸Ù Ù
ؤطر ÙÙ ÙØ¸ÙÙØ© عÙ
Ù â Ø¨Ø£Ùدا٠ÙKPI ÙØÙØ§Ø¬Ø² Ø£Ù
اÙ."},f2:{t:"Ù
تص٠بÙ
ÙØ¯Ø³Ù",d:"CRM ÙÙ
ÙØªØ¨ اÙÙ
ساعدة ÙØ§ÙØ¨Ø±ÙØ¯ ÙØ§ÙÙØ«Ø§Ø¦Ù ÙÙÙØ§Ø¹Ø¯ Ø§ÙØ¨ÙØ§ÙØ§Øª. تÙÙÙÙÙØ° Ø§ÙØ¥Ø¬Ø±Ø§Ø¡Ø§ØªØ ÙØ§ Ù
جرد Ø§ÙØªØ±Ø§ØÙا."},f3:{t:"Ù
ØÙÙÙ
ÙÙØ§Ø¨Ù ÙÙØªØ¯ÙÙÙ",d:"Ù٠إجراء Ù
سجÙ. تØÙÙ
Ø¨ØØ³Ø¨ Ø§ÙØ¯Ùر. Ù
ØªÙØ§ÙÙ Ù
ع GDPR. ÙÙØ¯Ø³Ø© سÙÙØ³Ø±ÙØ©."}},s2:{eyebrow:"Ø¨Ø¯ÙØ§Ø¦ÙØ ÙØ§ Ø¨Ø£Ø´ÙØ±",title:"Ø£ÙØ´Ø¦ Ù
ÙØ¸Ù AI ÙÙ Ø«ÙØ§Ø« Ø®Ø
811·Ùات.",st1:{t:"اختر Ø¯ÙØ±Ùا",d:"Ù
Ø¨ÙØ¹Ø§ØªØ دعÙ
Ø ØªØ³ÙÙÙØ Ù
اÙÙØ©Ø Ø¨ØØ« â Ø£Ù Ø¯ÙØ± Ù
خصص."},st2:{t:"اربط Ø£Ø¯ÙØ§ØªÙ",d:"Ù
ÙØµÙات بÙÙØ±Ø© ÙØ§ØØ¯Ø© Ø¥ÙÙ CRM ÙÙ
ÙØªØ¨ اÙÙ
ساعدة ÙØ§ÙÙØ«Ø§Ø¦Ù ÙÙÙØ§Ø¹Ø¯ Ø§ÙØ¨ÙØ§ÙØ§Øª."},st3:{t:"Ø§ÙØ´Ø± ÙÙØ±Ùر",d:"ÙØ¨Ø¯Ø£ Ù
ÙØ¸ÙÙ Ø§ÙØ¹Ù
Ù. راجع Ø§ÙØ¥Ø¬Ø±Ø§Ø¡Ø§ØªØ اضبط Ø§ÙØÙØ§Ø¬Ø²Ø Ø«Ù
ØªÙØ³Ùع."},cta:"جرب Ù
ÙØ´Ø¦ اÙÙ
ÙØ¸Ù"},s3:{eyebrow:"Ø£Ø¯ÙØ§Ø± Ø¬Ø§ÙØ²Ø© ÙÙÙØ´Ø±",title:"تعرÙ٠عÙ٠زÙ
ÙØ§Ø¡ AI.",explore:"Ø§Ø³ØªÙØ´Ù Ø§ÙØ¯Ùر"},s4:{eyebrow:"ÙÙ
اذا تختار اÙÙØ±Ù Infernex",title:"ÙØªØ§Ø¦Ø¬ ÙØ§Ø¨ÙØ© ÙÙÙÙØ§Ø³Ø ÙØ§ Ù
ÙØ§ÙÙØ³ Ù
Ø¸ÙØ±ÙØ©.",body:"ÙØ§ ÙÙØ´Ø± Ù
عاÙÙØ± Ù
ختÙÙØ©. Ø§ÙØ¹Ø§Ø¦Ø¯ اÙÙØ¹ÙÙ ÙØ¹ØªÙ
د عÙÙ Ø³ÙØ± عÙ
ÙÙ ÙØ£ØØ¬Ø§Ù
Ù."},s5:{eyebrow:"اÙÙ
ØØ±Ù ØªØØª Ø§ÙØºØ·Ø§Ø¡",title:"Ù
دعÙÙ
ب٠Infernex Core.",body:"ÙÙ Ù
ÙØ¸Ù AI ÙØ¹Ù
٠عÙÙ Infernex Core â Ù
ÙØ¬Ù٠ذÙÙ ÙÙØ²Ø¹ ÙÙ Ù
ÙÙ
Ø© Ø¥Ù٠اÙÙ
ØØ±Ùات اÙÙÙØªÙÙÙØ© أ٠اÙÙÙ
ÙÙ
ÙØ© Ø£Ù Ø§ÙØ«Ø±Ù
ÙØ¯ÙÙØ§Ù
ÙÙÙØ© Ø£Ù MiroFish.",cta:"Ø´Ø§ÙØ¯ Ù
ÙØµØ© اÙÙ
Ø·ÙÙØ±ÙÙ"},s6:{eyebrow:"ÙÙØ© Ø§ÙØ¹Ù
Ù Ù
ÙØ§Ø¨Ù Ø§ÙØ·Ø±ÙÙØ© اÙÙØ¯ÙÙ
Ø©",title:"Ù
ÙØµØ© ÙØ§ØØ¯Ø©. ÙØ§ عشر Ø£Ø¯ÙØ§Øª Ù
ÙÙÙØ©.",old:{h:"اÙÙ
ÙØ¯Ø³ Ø§ÙØªÙÙÙØ¯Ù",t:"براÙ
ج Ù
ØªØ¹Ø¯Ø¯Ø©Ø ØµÙØ§Ù
Ø¹Ø Ø£ØªÙ
تات ÙØ´Ø©.",i1:"SaaS Ù
ÙÙØµÙ ÙÙÙ ÙØ¸ÙÙØ©.",i2:"Ø³ÙØ§Ø³Ù Zapier تÙÙØ³Ø± Ù
ع تغÙÙØ± ÙØ§Ø¬Ùات API.",i3:"Ø£Ø¯ÙØ§Øª AI Ù
ÙØ¹Ø²ÙØ© Ø¨ÙØ§ Ø°Ø§ÙØ±Ø© ÙØ£Ø¹Ù
اÙÙ.",i4:"تتراÙÙ
Ø§ÙØªÙاÙÙÙ. تÙÙÙØ¯ Ø§ÙØÙÙÙ
Ø©."},new:{h:"ÙÙØ© عÙ
Ù Infernex Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",t:"ÙÙØ© عÙ
Ù ÙØ§ØØ¯Ø©Ø Ù
ÙØµØ© ÙØ§ØØ¯Ø©Ø ÙØ§ØªÙرة ÙØ§ØØ¯Ø©.",i1:"Ù
ÙØ¸ÙÙ AI ذاتÙÙÙ ÙÙÙ ÙØ¸ÙÙØ© عÙ
Ù.",i2:"Ù
ÙØµÙات Ù
ÙØØ¯Ø© Ø¥ÙÙ CRM ÙØ§ÙÙØ«Ø§Ø¦Ù ÙÙ
ÙØªØ¨ اÙÙ
ساعدة.",i3:"Ø°Ø§ÙØ±Ø© Ø·ÙÙÙØ© Ø§ÙØ£Ù
د ÙØ¨ÙØ§ÙØ§ØªÙ ÙØ¹Ù
ÙÙØ§ØªÙ.",i4:"إجراءات ÙØ§Ø¨ÙØ© ÙÙØªØ¯ÙÙÙØ ØÙÙÙ
Ø© ØØ³Ø¨ Ø§ÙØ¯ÙØ±Ø ÙØ§ØªÙرة ÙØ§ØØ¯Ø©."},cta:"Ø£ÙØ´Ø¦ Ø£ÙÙ Ù
ÙØ¸Ù Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù",microcopy:"خطة Ù
جاÙÙØ© Ø¥ÙÙ Ø§ÙØ£Ø¨Ø¯. رÙÙ Ù
ت٠شئت."},faq:{eyebrow:"Ø§ÙØ£Ø³Ø¦ÙØ© Ø§ÙØ´Ø§Ø¦Ø¹Ø©",title:"ÙÙ Ù
ا ØªØØªØ§Ø¬ Ù
Ø¹Ø±ÙØªÙ",subtitle:"Ø§ÙØ´Ø± ÙÙØ© عÙ
ÙÙ Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù ÙØÙÙÙÙ
ÙØ§ ÙÙØ³ÙØ¹ÙØ§ Ø¨Ø«ÙØ©.",q1:"Ù
ا Ù
ÙØ¸Ù Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù Ø¨Ø§ÙØªØØ¯ÙدØ",a1:"Ù
ÙØ¸Ù AI ÙÙ ÙÙÙÙ Ø°Ø§ØªÙ Ø¨Ø¯ÙØ± Ù
ØØ¯Ø¯ ÙØªØ¹ÙÙÙ
ات ÙØµÙاØÙات Ø£Ø¯ÙØ§Øª ÙØ°Ø§Ùرة Ø·ÙÙÙØ© Ø§ÙØ£Ù
د ÙØ£Ø¹Ù
اÙÙ.",q2:"ÙÙ
ÙØ³ØªØºØ±Ù ÙØ´Ø± Ø£ÙÙ Ù
ÙØ¸Ù AIØ",a2:"Ø£ÙÙ Ù
Ù 5 Ø¯ÙØ§Ø¦Ù. اختر ÙØ§ÙØ¨ÙØ§Ø ص٠Ù
صدر Ø§ÙØ¨ÙØ§ÙØ§ØªØ راجع اÙÙ
ÙÙ
Ø© ÙØ§ÙØµÙØ§ØÙØ§ØªØ Ø«Ù
Ø§ÙØ´Ø±.",q3:"ÙÙ Ø¨ÙØ§Ùات Ø´Ø±ÙØªÙ Ø¢Ù
ÙØ© ÙØ¬Ø§Ùزة ÙÙ GDPRØ",a3:"ÙØ¹Ù
. Ù٠إجراء Ù
سجÙÙ ÙÙ
عزÙÙ ÙÙÙ Ù
ÙØ¸Ù
Ø© ÙÙ
ØÙ
٠باعتÙ
اد بشر٠ÙÙ Ø§ÙØ¹Ù
ÙÙØ§Øª Ø§ÙØØ³Ø§Ø³Ø©.",q4:"ÙÙ Ø£ØØªÙظ Ø¨Ø§ÙØ³Ùطرة Ø§ÙØ¨Ø´Ø±ÙØ© عÙÙ Ù
ÙØ¸ÙÙ AIØ",a4:"دائÙ
ÙØ§. ØªØØ¯Ø¯ Ø§ÙØµÙاØÙات ÙÙ٠أداة ÙØ¥Ø¬Ø±Ø§Ø¡ ÙÙ
صدر Ø¨ÙØ§Ùات.",q5:"Ù
ا Ø§ÙØ£Ø¯Ùات ÙCRMs Ø§ÙØªÙ ÙÙ
ÙÙ Ø§ÙØ§ØªØµØ§Ù Ø¨ÙØ§Ø",a5:"Salesforce ÙHubSpot ÙZoho ÙPipedrive ÙSlack ÙGmail ÙGoogle Drive ÙNotion ÙÙÙØ§Ø¹Ø¯ Ø¨ÙØ§Ùات ÙStripe ÙZendesk ÙØ£Ùثر Ù
Ù 30 ØºÙØ±Ùا.",q6:"ÙÙÙ ÙØ¹Ù
Ù Ø§ÙØªØ³Ø¹ÙرØ",a6:"ابدأ Ù
جاÙÙØ§ باستخداÙ
Ù
ØØ¯Ùد. ØªÙØªØ Ø§ÙØ®Ø·Ø· اÙÙ
دÙÙØ¹Ø© Ù
Ø²ÙØ¯Ùا Ù
٠اÙÙ
ÙØ¸ÙÙÙ ÙØ§ÙتÙÙÙØ§Øª ÙØ§ÙÙ
ÙØµÙات.",q7:"ÙÙ ÙÙ
ÙÙÙ٠استخداÙ
LLM خاص Ø¨Ù Ø£Ù Ø§ÙØªÙجÙ٠إÙÙ ÙÙ
ÙØ°Ø¬ Ù
ØØ¯Ø¯Ø",a7:"ÙØ¹Ù
. ÙÙØ¬ÙÙ Infernex Core ÙÙ Ù
ÙÙ
Ø© Ø¥Ù٠اÙÙ
ØØ±Ù Ø§ÙØ£ÙÙØ£ Ø£Ù Ù
زÙÙØ¯Ù اÙÙ
ختار.",q8:"Ù
ا Ø§ÙØ¹Ø§Ø¦Ø¯ اÙÙØ§Ùع٠اÙÙ
تÙÙØ¹Ø",a8:"تÙÙØ¯ اÙÙØ±Ù Ø¹Ø§Ø¯Ø©Ù Ø¨Ø§ÙØ®Ùاض 40â70Ùª ÙÙ Ø§ÙØ¹Ø¨Ø¡ اÙÙ
ØªÙØ±Ø± Ø®ÙØ§Ù Ø£ÙÙ 60 ÙÙÙ
ÙØ§.",cta:"ØªØØ¯Ø« Ø¥ÙÙ ÙØ±ÙÙÙØ§"}}}};for(const[e,t]of Object.entries(Xu))Ve[e]=t;const em={en:{eyebrow:"Positioning",title:"A complete AI workforce built for your company.",subtitle:"Salesforce, Microsoft Copilot and ServiceNow automate specific workflows. Infernex deploys autonomous AI employees that collaborate across your entire organization.",table:{feature:"Feature",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"Core Product",sf:"CRM + customer data",ms:"Productivity suite + enterprise operating system",sn:"ITSM + enterprise workflows",ix:"Autonomous multi-department AI Workforce"},users:{h:"Primary Users",sf:"Sales teams",ms:"Microsoft 365 employees",sn:"IT and operations teams",ix:"Entire company + executives"},agents:{h:"AI Agents",sf:"Specialized agents inside the Salesforce ecosystem",ms:"Assistants connected to the Microsoft environment",sn:"Workflow automation agents",ix:"Specialized AI employees collaborating together"}
811,dep:{h:"Dependency",sf:"Highly dependent on the Salesforce ecosystem",ms:"Highly dependent on the Microsoft ecosystem",sn:"Highly dependent on the ServiceNow ecosystem",ix:"Works independently with your existing company tools"},models:{h:"AI Models",sf:"Mainly their ecosystem",ms:"OpenAI + Microsoft infrastructure",sn:"Integrated models",ix:"Intelligent multi-LLM routing optimizing cost and performance"},deploy:{h:"Deployment",sf:"Long enterprise deployment",ms:"License-based deployment",sn:"Heavy enterprise implementation",ix:"Fast SaaS deployment"},size:{h:"Company Size",sf:"Enterprise focused",ms:"Medium to enterprise",sn:"Large enterprises",ix:"From startups and SMBs to global enterprises"},market:{h:"Marketplace",sf:"Salesforce ecosystem",ms:"AppSource ecosystem",sn:"ServiceNow Store",ix:"Universal AI Employee Marketplace"}},os:{eyebrow:"The Difference",title:"The AI Workforce OS",body:"The smallest or biggest companies don't need another chatbot. They need a complete AI digital workforce."},flow:{eyebrow:"AI Workforce Example",title:"Sales Growth Team",subtitle:"Multiple AI employees connected together through the Infernex AI Workforce OS.",orchestrator:"Orchestrator",r1:{t:"AI SDR",d:"Finds and qualifies prospects"},r2:{t:"Marketing Agent",d:"Creates campaigns and content"},r3:{t:"Sales Agent",d:"Manages opportunities"},r4:{t:"CFO Agent",d:"Analyzes revenue and profitability"},r5:{t:"Customer Success Agent",d:"Retains customers"},r6:{t:"AI COO",d:"Coordinates the entire AI workforce"}},final:"You don't manage a chatbot. You manage an entire AI company.",cta:{primary:"Deploy Your AI Workforce",secondary:"Explore AI Employees"},links:{aria:"Learn more",workforce:"AI Workforce",marketplace:"AI Employee Marketplace",pricing:"Pricing",docs:"Documentation"}},fr:{eyebrow:"Positionnement",title:"Une équipe IA complète conçue pour votre entreprise.",subtitle:"Salesforce, Microsoft Copilot et ServiceNow automatisent certains workflows. Infernex déploie des employés IA autonomes qui collaborent dans toute votre organisation.",table:{feature:"Critère",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"Produit principal",sf:"CRM + données clients",ms:"Suite bureautique + système dâexploitation dâentreprise",sn:"ITSM + workflows dâentreprise",ix:"Force de travail IA autonome multi-départements"},users:{h:"Utilisateurs cibles",sf:"Ãquipes commerciales",ms:"Employés Microsoft 365",sn:"Ãquipes IT et opérations",ix:"Toute lâentreprise + les dirigeants"},agents:{h:"Agents IA",sf:"Agents spécialisés au sein de lâécosystème Salesforce",ms:"Assistants connectés à lâenvironnement Microsoft",sn:"Agents dâautomatisation de workflows",ix:"Employés IA spécialisés qui collaborent entre eux"},dep:{h:"Dépendance",sf:"Fortement dépendant de lâécosystème Salesforce",ms:"Fortement dépendant de lâécosystème Microsoft",sn:"Fortement dépendant de lâécosystème ServiceNow",ix:"Fonctionne indépendamment avec vos outils existants"},models:{h:"Modèles IA",sf:"Principalement leur écosystème",ms:"OpenAI + infrastructure Microsoft",sn:"Modèles intégrés",ix:"Routage multi-LLM intelligent optimisant coût et performance"},deploy:{h:"Déploiement",sf:"Déploiement entreprise long",ms:"Déploiement par licences",sn:"Implémentation lourde",ix:"Déploiement SaaS rapide"},size:{h:"Taille dâentreprise",sf:"Orienté grandes entreprises",ms:"Moyennes et grandes entreprises",sn:"Grandes entreprises",ix:"Des startups et PME aux multinationales"},market:{h:"Marketplace",sf:"Ãcosystème Salesforce",ms:"Ãcosystème AppSource",sn:"ServiceNow Store",ix:"Marketplace universelle dâemployés IA"}},os:{eyebrow:"La différence",title:"LâOS de la force de travail IA",body:"Les plus petites comme les plus grandes entreprises nâont pas besoin dâun chatbot de plus. Elles ont besoin dâune véritable force de travail IA digitale."},flow:{eyebrow:"Exemple de force de travail IA",title:"Ãquipe de croissance commerciale",subtitle:"Plusieurs employés IA connectés ensemble via lâOS Infernex AI Workforce.",orchestrator:"Orchestrateur",r1:{t:"SDR IA",d:"Trouve et qualifie les prospects"},r2:{t:"Agent Marketing",d:"Crée des campagnes et du contenu"},r3:{t:"Agent Commercial",d:"Gère les opportunités"},r4:{t:"Agent CFO",d:"Analyse le revenu et la rentabilité"},r5:{t:"Agent Customer Success",d:"Fidélise les clients"},r6:{t:"COO IA",d:"Coordonne lâensemble de la force de travail IA"}},final:"Vous ne pilotez pas un chatbot. Vous pilotez une entreprise IA à part entière.",cta:{primary:"Déployer votre force de travail IA",secondary:"Explorer les employés IA"},links:{aria:"En savoir plus",workforce:"Force de travail IA",marketplace:"Marketplace des employés IA",pricing:"Tarifs",docs:"Documentation"}},es:{eyebrow:"Posicionamiento",title:"Una fuerza laboral IA completa para tu empresa.",subt
811itle:"Salesforce, Microsoft Copilot y ServiceNow automatizan flujos especÃficos. Infernex despliega empleados IA autónomos que colaboran en toda tu organización.",table:{feature:"Criterio",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"Producto principal",sf:"CRM + datos de clientes",ms:"Suite de productividad + sistema operativo empresarial",sn:"ITSM + flujos empresariales",ix:"Fuerza laboral IA autónoma multi-departamento"},users:{h:"Usuarios objetivo",sf:"Equipos de ventas",ms:"Empleados Microsoft 365",sn:"Equipos IT y operaciones",ix:"Toda la empresa + directivos"},agents:{h:"Agentes IA",sf:"Agentes especializados en el ecosistema Salesforce",ms:"Asistentes conectados al entorno Microsoft",sn:"Agentes de automatización de flujos",ix:"Empleados IA especializados que colaboran entre sÃ"},dep:{h:"Dependencia",sf:"Muy dependiente del ecosistema Salesforce",ms:"Muy dependiente del ecosistema Microsoft",sn:"Muy dependiente del ecosistema ServiceNow",ix:"Funciona de forma independiente con tus herramientas actuales"},models:{h:"Modelos IA",sf:"Principalmente su ecosistema",ms:"OpenAI + infraestructura Microsoft",sn:"Modelos integrados",ix:"Enrutamiento multi-LLM inteligente que optimiza coste y rendimiento"},deploy:{h:"Despliegue",sf:"Despliegue empresarial largo",ms:"Despliegue por licencias",sn:"Implementación empresarial pesada",ix:"Despliegue SaaS rápido"},size:{h:"Tamaño de empresa",sf:"Orientado a grandes empresas",ms:"Medianas a grandes",sn:"Grandes empresas",ix:"Desde startups y pymes hasta multinacionales"},market:{h:"Marketplace",sf:"Ecosistema Salesforce",ms:"Ecosistema AppSource",sn:"ServiceNow Store",ix:"Marketplace universal de empleados IA"}},os:{eyebrow:"La diferencia",title:"El SO de la fuerza laboral IA",body:"Ni las empresas más pequeñas ni las más grandes necesitan otro chatbot. Necesitan una fuerza laboral IA digital completa."},flow:{eyebrow:"Ejemplo de fuerza laboral IA",title:"Equipo de crecimiento comercial",subtitle:"Múltiples empleados IA conectados a través del SO Infernex AI Workforce.",orchestrator:"Orquestador",r1:{t:"SDR IA",d:"Encuentra y cualifica prospectos"},r2:{t:"Agente de Marketing",d:"Crea campañas y contenido"},r3:{t:"Agente de Ventas",d:"Gestiona las oportunidades"},r4:{t:"Agente CFO",d:"Analiza ingresos y rentabilidad"},r5:{t:"Agente Customer Success",d:"Fideliza a los clientes"},r6:{t:"COO IA",d:"Coordina toda la fuerza laboral IA"}},final:"No gestionas un chatbot. Gestionas toda una empresa IA.",cta:{primary:"Despliega tu fuerza laboral IA",secondary:"Explora los empleados IA"},links:{aria:"Saber más",workforce:"Fuerza laboral IA",marketplace:"Marketplace de empleados IA",pricing:"Precios",docs:"Documentación"}},de:{eyebrow:"Positionierung",title:"Eine komplette KI-Belegschaft für Ihr Unternehmen.",subtitle:"Salesforce, Microsoft Copilot und ServiceNow automatisieren einzelne Workflows. Infernex stellt autonome KI-Mitarbeiter bereit, die im gesamten Unternehmen zusammenarbeiten.",table:{feature:"Kriterium",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"Kernprodukt",sf:"CRM + Kundendaten",ms:"Produktivitätssuite + Unternehmens-Betriebssystem",sn:"ITSM + Unternehmens-Workflows",ix:"Autonome bereichsübergreifende KI-Belegschaft"},users:{h:"Zielnutzer",sf:"Vertriebsteams",ms:"Microsoft-365-Mitarbeitende",sn:"IT- und Operations-Teams",ix:"Gesamtes Unternehmen + Führungskräfte"},agents:{h:"KI-Agenten",sf:"Spezialisierte Agenten im Salesforce-Ãkosystem",ms:"Assistenten im Microsoft-Umfeld",sn:"Workflow-Automatisierungsagenten",ix:"Spezialisierte KI-Mitarbeiter, die zusammenarbeiten"},dep:{h:"Abhängigkeit",sf:"Stark vom Salesforce-Ãkosystem abhängig",ms:"Stark vom Microsoft-Ãkosystem abhängig",sn:"Stark vom ServiceNow-Ãkosystem abhängig",ix:"Arbeitet unabhängig mit Ihren bestehenden Tools"},models:{h:"KI-Modelle",sf:"Hauptsächlich eigenes Ãkosystem",ms:"OpenAI + Microsoft-Infrastruktur",sn:"Integrierte Modelle",ix:"Intelligentes Multi-LLM-Routing für Kosten und Leistung"},deploy:{h:"Einführung",sf:"Lange Enterprise-Einführung",ms:"Lizenzbasierte Einführung",sn:"Aufwendige Enterprise-Implementierung",ix:"Schnelle SaaS-Einführung"},size:{h:"UnternehmensgröÃe",sf:"Fokus GroÃunternehmen",ms:"Mittelstand bis Konzern",sn:"GroÃe Unternehmen",ix:"Vom Start-up und KMU bis zum Weltkonzern"},market:{h:"Marketplace",sf:"Salesforce-Ãkosystem",ms:"AppSource-Ãkosystem",sn:"ServiceNow Store",ix:"Universeller KI-Mitarbeiter-Marketplace"}},os:{eyebrow:"Der Unterschied",title:"Das KI-Workforce-OS",body:"Weder kleine noch groÃe Unternehmen brauchen einen weiteren Chatbot. Sie brauchen eine vollständige digitale KI-Belegschaft."},flow:{eyebrow:"KI-Workforce-Beispiel",title:"Sales-Growth-Team",subtitle:"Mehrere KI-Mitarbeiter verbunden über das Infernex-AI-Workforce-OS.",orchestrator:"Orchestrator",r1:{t:"KI-SDR",d:"Findet und qualifiziert Interessenten"},r2:{t:"Marketing-Agent",d:"Erstellt Kampagnen und Inhalte"},r3:{t:"Sales-Agent",d:"Betreut Opportunities"}
811,r4:{t:"CFO-Agent",d:"Analysiert Umsatz und Profitabilität"},r5:{t:"Customer-Success-Agent",d:"Bindet Kunden langfristig"},r6:{t:"KI-COO",d:"Koordiniert die gesamte KI-Belegschaft"}},final:"Sie steuern keinen Chatbot. Sie steuern ein ganzes KI-Unternehmen.",cta:{primary:"KI-Belegschaft einsetzen",secondary:"KI-Mitarbeiter entdecken"},links:{aria:"Mehr erfahren",workforce:"KI-Workforce",marketplace:"KI-Mitarbeiter-Marketplace",pricing:"Preise",docs:"Dokumentation"}},pt:{eyebrow:"Posicionamento",title:"Uma força de trabalho IA completa para a sua empresa.",subtitle:"Salesforce, Microsoft Copilot e ServiceNow automatizam fluxos especÃficos. A Infernex implanta funcionários IA autónomos que colaboram em toda a sua organização.",table:{feature:"Critério",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"Produto principal",sf:"CRM + dados de clientes",ms:"Suite de produtividade + sistema operativo empresarial",sn:"ITSM + fluxos empresariais",ix:"Força de trabalho IA autónoma multi-departamento"},users:{h:"Utilizadores-alvo",sf:"Equipas de vendas",ms:"Colaboradores Microsoft 365",sn:"Equipas de TI e operações",ix:"Toda a empresa + direção"},agents:{h:"Agentes IA",sf:"Agentes especializados no ecossistema Salesforce",ms:"Assistentes ligados ao ambiente Microsoft",sn:"Agentes de automatização de fluxos",ix:"Funcionários IA especializados que colaboram entre si"},dep:{h:"Dependência",sf:"Muito dependente do ecossistema Salesforce",ms:"Muito dependente do ecossistema Microsoft",sn:"Muito dependente do ecossistema ServiceNow",ix:"Funciona de forma independente com as suas ferramentas atuais"},models:{h:"Modelos IA",sf:"Principalmente o seu ecossistema",ms:"OpenAI + infraestrutura Microsoft",sn:"Modelos integrados",ix:"Encaminhamento multi-LLM inteligente que otimiza custo e desempenho"},deploy:{h:"Implementação",sf:"Implementação empresarial longa",ms:"Implementação por licenças",sn:"Implementação empresarial pesada",ix:"Implementação SaaS rápida"},size:{h:"Dimensão da empresa",sf:"Focado em grandes empresas",ms:"Médias a grandes",sn:"Grandes empresas",ix:"De startups e PME a multinacionais"},market:{h:"Marketplace",sf:"Ecossistema Salesforce",ms:"Ecossistema AppSource",sn:"ServiceNow Store",ix:"Marketplace universal de funcionários IA"}},os:{eyebrow:"A diferença",title:"O SO da força de trabalho IA",body:"Nem as menores nem as maiores empresas precisam de mais um chatbot. Precisam de uma força de trabalho digital IA completa."},flow:{eyebrow:"Exemplo de força de trabalho IA",title:"Equipa de crescimento comercial",subtitle:"Vários funcionários IA ligados através do SO Infernex AI Workforce.",orchestrator:"Orquestrador",r1:{t:"SDR IA",d:"Encontra e qualifica prospetos"},r2:{t:"Agente de Marketing",d:"Cria campanhas e conteúdo"},r3:{t:"Agente de Vendas",d:"Gere oportunidades"},r4:{t:"Agente CFO",d:"Analisa receita e rentabilidade"},r5:{t:"Agente Customer Success",d:"Fideliza clientes"},r6:{t:"COO IA",d:"Coordena toda a força de trabalho IA"}},final:"Não gere um chatbot. Gere uma empresa IA inteira.",cta:{primary:"Implante a sua força de trabalho IA",secondary:"Explore os funcionários IA"},links:{aria:"Saber mais",workforce:"Força de trabalho IA",marketplace:"Marketplace de funcionários IA",pricing:"Preços",docs:"Documentação"}},zh:{eyebrow:"å®ä½",title:"为æ¨çå
¬å¸æé ç宿´ AI åå·¥éä¼ã",subtitle:"SalesforceãMicrosoft Copilot å ServiceNow åªèªå¨åç¹å®å·¥ä½æµãInfernex é¨ç½²èªæ²»ç AI åå·¥ï¼è·¨æ´ä¸ªç»ç»åä½ã",table:{feature:"对æ¯é¡¹",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"æ ¸å¿äº§å",sf:"CRM + å®¢æ·æ°æ®",ms:"ç产åå¥ä»¶ + ä¼ä¸æä½ç³»ç»",sn:"ITSM + ä¼ä¸æµç¨",ix:"è·¨é¨é¨èªæ²» AI åå·¥éä¼"},users:{h:"主è¦ç¨æ·",sf:"éå®å¢é",ms:"Microsoft 365 åå·¥",sn:"IT ä¸è¿è¥å¢é",ix:"æ´ä¸ªå
¬å¸ + 管çå±"},agents:{h:"AI æºè½ä½",sf:"Salesforce çæå
çä¸ä¸æºè½ä½",ms:"æ¥å
¥ Microsoft ç¯å¢ç婿",sn:"工使µèªå¨åæºè½ä½",ix:"å¤ä¸ªä¸ä¸ AI åå·¥ååå·¥ä½"},dep:{h:"ä¾èµåº¦",sf:"é«åº¦ä¾èµ Salesforce çæ",ms:"é«åº¦ä¾èµ Microsoft çæ",sn:"é«åº¦ä¾èµ ServiceNow çæ",ix:"ç¬ç«è¿è¡ï¼å
¼å®¹æ¨ç°æå·¥å
·"},models:{h:"AI 模å",sf:"ä¸»è¦æ¥èªèªå®¶çæ",ms:"OpenAI + Microsoft åºç¡è®¾æ½",sn:"å
ç½®éææ¨¡å",ix:"å¤ LLM æºè½è·¯ç±ï¼å
¼é¡¾ææ¬ä¸æ§è½"},deploy:{h:"é¨ç½²æ¹å¼",sf:"漫é¿çä¼ä¸çº§é¨ç½²",ms:"æè®¸å¯è¯é¨ç½²",sn:"é度ä¼ä¸çº§å®æ½",ix:"å¿«é SaaS é¨ç½²"},size:{h:"ä¼ä¸è§æ¨¡",sf:"é¢å大åä¼ä¸",ms:"ä¸åå°å¤§åä¼ä¸",sn:"大åä¼ä¸",ix:"ä»åååä¸å°ä¼ä¸å°å
¨çéå¢"},market:{h:"å¸åºçæ",sf:"Salesforce çæ",ms:"AppSource çæ",sn:"ServiceNow Store",ix:"éç¨ AI åå·¥å¸åº"}},os:{eyebrow:"差弿å¨",title:"AI åå·¥æä½ç³»ç»",body:"æ 论å°ä¼ä¸è¿æ¯å¤§éå¢ï¼éè¦çé½ä¸æ¯åä¸ä¸ªè天æºå¨äººï¼èæ¯ä¸æ¯å®æ´çæ°åå AI åå·¥éä¼ã"},flow:{eyebrow:"AI åå·¥éä¼ç¤ºä¾",title:"éå®å¢é¿å¢é",subtitle:"éè¿ Infernex AI Workforce OSï¼å¤ä¸ª AI å工紧å¯ååã",orchestrator:"ç¼æè
",r1:{t:"AI SDR",d:"åç°å¹¶çéæ½å¨å®¢æ·"},r2:{t:"è¥éæºè½ä½",d:"å建活å¨ä¸å
容"},r3:{t:"é宿ºè½ä½",d:"管çåæº"}
811,r4:{t:"CFO æºè½ä½",d:"åææ¶å
¥ä¸çå©"},r5:{t:"å®¢æ·æåæºè½ä½",d:"çä½å®¢æ·"},r6:{t:"AI COO",d:"ç»ç¹æ´ä¸ª AI åå·¥éä¼"}},final:"æ¨ç®¡ççä¸åæ¯ä¸ä¸ªè天æºå¨äººï¼èæ¯ä¸æ´å®¶ AI å
¬å¸ã",cta:{primary:"é¨ç½²æ¨ç AI åå·¥éä¼",secondary:"æµè§ AI åå·¥"},links:{aria:"äºè§£æ´å¤",workforce:"AI åå·¥éä¼",marketplace:"AI åå·¥å¸åº",pricing:"ä»·æ ¼",docs:"ææ¡£"}},ja:{eyebrow:"ãã¸ã·ã§ãã³ã°",title:"伿¥ã®ããã«æ§ç¯ãããå®å
¨ãª AI ã¯ã¼ã¯ãã©ã¼ã¹ã§ãã",subtitle:"SalesforceãMicrosoft CopilotãServiceNow ã¯ç¹å®ã®ã¯ã¼ã¯ããã¼ãèªååãã¾ããInfernex ã¯çµç¹å
¨ä½ã§é£æºããèªå¾ç㪠AI 徿¥å¡ãå±éãã¾ãã",table:{feature:"é
ç®",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"主è¦è£½å",sf:"CRM + 顧客ãã¼ã¿",ms:"çç£æ§ã¹ã¤ã¼ã + 伿¥åã OS",sn:"ITSM + 伿¥ã¯ã¼ã¯ããã¼",ix:"é¨é横æã§åãèªå¾å AI ã¯ã¼ã¯ãã©ã¼ã¹"},users:{h:"主ãªã¦ã¼ã¶ã¼",sf:"å¶æ¥ãã¼ã ",ms:"Microsoft 365 ã®å¾æ¥å¡",sn:"ITã»ãªãã¬ã¼ã·ã§ã³é¨é",ix:"å
¨ç¤¾å¡ã¨çµå¶å±¤"},agents:{h:"AI ã¨ã¼ã¸ã§ã³ã",sf:"Salesforce ã¨ã³ã·ã¹ãã å
ã®å°éã¨ã¼ã¸ã§ã³ã",ms:"Microsoft ç°å¢ã«æ¥ç¶ãããã¢ã·ã¹ã¿ã³ã",sn:"ã¯ã¼ã¯ããã¼èªååã¨ã¼ã¸ã§ã³ã",ix:"飿ºãã¦åãå°é AI 徿¥å¡"},dep:{h:"ä¾å度",sf:"Salesforce ã¨ã³ã·ã¹ãã ã¸å¼·ãä¾å",ms:"Microsoft ã¨ã³ã·ã¹ãã ã¸å¼·ãä¾å",sn:"ServiceNow ã¨ã³ã·ã¹ãã ã¸å¼·ãä¾å",ix:"æ¢åãã¼ã«ã¨ç¬ç«ãã¦åä½"},models:{h:"AI ã¢ãã«",sf:"主ã«èªç¤¾ã¨ã³ã·ã¹ãã ",ms:"OpenAI + Microsoft ã¤ã³ãã©",sn:"çµ±åã¢ãã«",ix:"ã³ã¹ãã¨æ§è½ãæé©åãããã«ã LLM ã«ã¼ãã£ã³ã°"},deploy:{h:"å°å
¥",sf:"é·æã®ã¨ã³ã¿ã¼ãã©ã¤ãºå°å
¥",ms:"ã©ã¤ã»ã³ã¹ãã¼ã¹å°å
¥",sn:"ééç´ã®ã¨ã³ã¿ã¼ãã©ã¤ãºå°å
¥",ix:"é«é㪠SaaS å°å
¥"},size:{h:"伿¥è¦æ¨¡",sf:"ã¨ã³ã¿ã¼ãã©ã¤ãºä¸å¿",ms:"ä¸å
ã大伿¥",sn:"大伿¥",ix:"ã¹ã¿ã¼ãã¢ããã»ä¸å°ããä¸çç大伿¥ã¾ã§"},market:{h:"ãã¼ã±ãããã¬ã¤ã¹",sf:"Salesforce ã¨ã³ã·ã¹ãã ",ms:"AppSource ã¨ã³ã·ã¹ãã ",sn:"ServiceNow Store",ix:"ã¦ããã¼ãµã« AI 徿¥å¡ãã¼ã±ãããã¬ã¤ã¹"}},os:{eyebrow:"éã",title:"AI ã¯ã¼ã¯ãã©ã¼ã¹ OS",body:"å°ããªä¼æ¥ã巍大伿¥ããããä¸ã¤ã®ãã£ãããããã§ã¯ãªããå®å
¨ãªãã¸ã¿ã« AI ã¯ã¼ã¯ãã©ã¼ã¹ãå¿
è¦ã§ãã"},flow:{eyebrow:"AI ã¯ã¼ã¯ãã©ã¼ã¹ã®ä¾",title:"ã»ã¼ã«ã¹æé·ãã¼ã ",subtitle:"Infernex AI Workforce OS ã§è¤æ°ã® AI 徿¥å¡ã飿ºãã¾ãã",orchestrator:"ãªã¼ã±ã¹ãã¬ã¼ã¿ã¼",r1:{t:"AI SDR",d:"è¦è¾¼ã¿å®¢ãçºæãè©ä¾¡"},r2:{t:"ãã¼ã±ãã£ã³ã°ã¨ã¼ã¸ã§ã³ã",d:"ãã£ã³ãã¼ã³ã¨ã³ã³ãã³ãã使"},r3:{t:"ã»ã¼ã«ã¹ã¨ã¼ã¸ã§ã³ã",d:"åè«ã管ç"},r4:{t:"CFO ã¨ã¼ã¸ã§ã³ã",d:"売ä¸ã¨åçæ§ãåæ"},r5:{t:"ã«ã¹ã¿ãã¼ãµã¯ã»ã¹ã¨ã¼ã¸ã§ã³ã",d:"顧客ãç¶æ"},r6:{t:"AI COO",d:"AI ã¯ã¼ã¯ãã©ã¼ã¹å
¨ä½ãçµ±æ¬"}},final:"ããªããåããã®ã¯ãã£ãããããã§ã¯ãªããã¾ããã¨ä¸ã¤ã® AI ã«ã³ããã¼ã§ãã",cta:{primary:"AI ã¯ã¼ã¯ãã©ã¼ã¹ãå±éãã",secondary:"AI 徿¥å¡ãè¦ã"},links:{aria:"詳細ãè¦ã",workforce:"AI ã¯ã¼ã¯ãã©ã¼ã¹",marketplace:"AI 徿¥å¡ãã¼ã±ãããã¬ã¤ã¹",pricing:"æé",docs:"ããã¥ã¡ã³ã"}},hi:{eyebrow:"पà¥à¤à¤¼à¤¿à¤¶à¤¨à¤¿à¤à¤",title:"à¤à¤ªà¤à¥ à¤à¤à¤ªà¤¨à¥ à¤à¥ लिठबनॠसà¤à¤ªà¥à¤°à¥à¤£ AI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸à¥¤",subtitle:"Salesforce, Microsoft Copilot à¤à¤° ServiceNow विशिषà¥à¤ वरà¥à¤à¤«à¤¼à¥à¤²à¥ à¤à¥ सà¥à¤µà¤à¤¾à¤²à¤¿à¤¤ à¤à¤°à¤¤à¥ हà¥à¤à¥¤ Infernex à¤à¤¸à¥ सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¤¤à¤¾ हॠà¤à¥ पà¥à¤°à¥ सà¤à¤à¤ न मà¥à¤ मिलà¤à¤° à¤à¤¾à¤® à¤à¤°à¤¤à¥ हà¥à¤à¥¤",table:{feature:"मापदà¤à¤¡",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"}
811,rows:{product:{h:"मà¥à¤à¥à¤¯ à¤à¤¤à¥à¤ªà¤¾à¤¦",sf:"CRM + à¤à¥à¤°à¤¾à¤¹à¤ डà¥à¤à¤¾",ms:"पà¥à¤°à¥à¤¡à¤à¥à¤à¤¿à¤µà¤¿à¤à¥ सà¥à¤ + à¤à¤à¤à¤°à¤ªà¥à¤°à¤¾à¤à¤à¤¼ à¤à¤ªà¤°à¥à¤à¤¿à¤à¤ सिसà¥à¤à¤®",sn:"ITSM + à¤à¤à¤à¤°à¤ªà¥à¤°à¤¾à¤à¤à¤¼ वरà¥à¤à¤«à¤¼à¥à¤²à¥",ix:"बहà¥-विà¤à¤¾à¤à¥à¤¯ सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸"},users:{h:"मà¥à¤à¥à¤¯ à¤à¤ªà¤¯à¥à¤à¤à¤°à¥à¤¤à¤¾",sf:"सà¥à¤²à¥à¤¸ à¤à¥à¤®",ms:"Microsoft 365 à¤à¤°à¥à¤®à¤à¤¾à¤°à¥",sn:"IT à¤à¤° à¤à¤ªà¤°à¥à¤¶à¤à¤¸ à¤à¥à¤®",ix:"पà¥à¤°à¥ à¤à¤à¤ªà¤¨à¥ + à¤à¤¾à¤°à¥à¤¯à¤à¤¾à¤°à¥"},agents:{h:"AI à¤à¤à¥à¤à¤",sf:"Salesforce à¤à¤à¥à¤¸à¤¿à¤¸à¥à¤à¤® à¤à¥ विशà¥à¤· à¤à¤à¥à¤à¤",ms:"Microsoft वातावरण सॠà¤à¥à¤¡à¤¼à¥ à¤
सिसà¥à¤à¥à¤à¤",sn:"वरà¥à¤à¤«à¤¼à¥à¤²à¥ à¤à¤à¥à¤®à¥à¤¶à¤¨ à¤à¤à¥à¤à¤",ix:"मिलà¤à¤° à¤à¤¾à¤® à¤à¤°à¤¨à¥ वालॠविशà¥à¤· AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥"},dep:{h:"निरà¥à¤à¤°à¤¤à¤¾",sf:"Salesforce à¤à¤à¥à¤¸à¤¿à¤¸à¥à¤à¤® पर बहà¥à¤¤ निरà¥à¤à¤°",ms:"Microsoft à¤à¤à¥à¤¸à¤¿à¤¸à¥à¤à¤® पर बहà¥à¤¤ निरà¥à¤à¤°",sn:"ServiceNow à¤à¤à¥à¤¸à¤¿à¤¸à¥à¤à¤® पर बहà¥à¤¤ निरà¥à¤à¤°",ix:"à¤à¤ªà¤à¥ मà¥à¤à¥à¤¦à¤¾ à¤à¥à¤²à¥à¤¸ à¤à¥ साथ सà¥à¤µà¤¤à¤à¤¤à¥à¤° रà¥à¤ª सॠà¤à¤¾à¤® à¤à¤°à¤¤à¤¾ हà¥"},models:{h:"AI मà¥à¤¡à¤²",sf:"मà¥à¤à¥à¤¯à¤¤à¤ à¤à¤¨à¤à¤¾ à¤
पना à¤à¤à¥à¤¸à¤¿à¤¸à¥à¤à¤®",ms:"OpenAI + Microsoft à¤à¤à¤«à¥à¤°à¤¾à¤¸à¥à¤à¥à¤°à¤à¥à¤à¤°",sn:"à¤à¤à¥à¤à¥à¤¤ मà¥à¤¡à¤²",ix:"लाà¤à¤¤ à¤à¤° पà¥à¤°à¤¦à¤°à¥à¤¶à¤¨ à¤à¥ à¤
नà¥à¤à¥à¤²à¤¿à¤¤ à¤à¤°à¤¨à¥ वाला बहà¥-LLM रà¥à¤à¤¿à¤à¤"},deploy:{h:"डिपà¥à¤²à¥à¤¯à¤®à¥à¤à¤",sf:"लà¤à¤¬à¤¾ à¤à¤à¤à¤°à¤ªà¥à¤°à¤¾à¤à¤à¤¼ डिपà¥à¤²à¥à¤¯à¤®à¥à¤à¤",ms:"लाà¤à¤¸à¥à¤à¤¸ à¤à¤§à¤¾à¤°à¤¿à¤¤ डिपà¥à¤²à¥à¤¯à¤®à¥à¤à¤",sn:"à¤à¤¾à¤°à¥ à¤à¤à¤à¤°à¤ªà¥à¤°à¤¾à¤à¤à¤¼ à¤à¤®à¥à¤ªà¥à¤²à¥à¤®à¥à¤à¤à¥à¤¶à¤¨",ix:"तà¥à¤à¤¼ SaaS डिपà¥à¤²à¥à¤¯à¤®à¥à¤à¤"},size:{h:"à¤à¤à¤ªà¤¨à¥ à¤à¤¾ à¤à¤à¤¾à¤°",sf:"बड़ॠà¤à¤à¤ªà¤¨à¤¿à¤¯à¥à¤ पर à¤à¥à¤à¤¦à¥à¤°à¤¿à¤¤",ms:"मधà¥à¤¯à¤® सॠबड़à¥",sn:"बड़ॠà¤à¤à¤ªà¤¨à¤¿à¤¯à¤¾à¤",ix:"सà¥à¤à¤¾à¤°à¥à¤à¤
प à¤à¤° SMB सॠलà¥à¤à¤° बहà¥à¤°à¤¾à¤·à¥à¤à¥à¤°à¥à¤¯ à¤à¤à¤ªà¤¨à¤¿à¤¯à¥à¤ तà¤"},market:{h:"मारà¥à¤à¥à¤à¤ªà¥à¤²à¥à¤¸",sf:"Salesforce à¤à¤à¥à¤¸à¤¿à¤¸à¥à¤à¤®",ms:"AppSource à¤à¤à¥à¤¸à¤¿à¤¸à¥à¤à¤®",sn:"ServiceNow Store",ix:"यà¥à¤¨à¤¿à¤µà¤°à¥à¤¸à¤² AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ मारà¥à¤à¥à¤à¤ªà¥à¤²à¥à¤¸"}},os:{eyebrow:"फ़रà¥à¤",title:"AI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸ OS",body:"à¤à¥à¤à¥ हà¥à¤ या बड़à¥, à¤à¤à¤ªà¤¨à¤¿à¤¯à¥à¤ à¤à¥ à¤à¤ à¤à¤° à¤à¥à¤à¤¬à¥à¤ नहà¥à¤ à¤à¤¾à¤¹à¤¿à¤à¥¤ à¤à¤¨à¥à¤¹à¥à¤ पà¥à¤°à¥ डिà¤à¤¿à¤à¤² AI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸ à¤à¤¾à¤¹à¤¿à¤à¥¤"},flow:{eyebrow:"AI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸ à¤à¤¾ à¤à¤¦à¤¾à¤¹à¤°à¤£",title:"सà¥à¤²à¥à¤¸ à¤à¥à¤°à¥à¤¥ à¤à¥à¤®",subtitle:"Infernex AI Workforce OS à¤à¥ à¤à¤¼à¤°à¤¿à¤ à¤à¤ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¤ साथ à¤à¥à¤¡à¤¼à¤¤à¥ हà¥à¤à¥¤",orchestrator:"à¤à¤°à¥à¤à¥à¤¸à¥à¤à¥à¤°à¥à¤à¤°",r1:{t:"AI SDR",d:"पà¥à¤°à¥à¤¸à¥à¤ªà¥à¤à¥à¤ à¤à¥à¤à¤¤à¤¾ à¤à¤° à¤à¥à¤µà¤¾à¤²à¤¿à¤«à¤¾à¤ à¤à¤°à¤¤à¤¾ हà¥"},r2:{t:"मारà¥à¤à¥à¤à¤¿à¤à¤ à¤à¤à¥à¤à¤",d:"à¤à¥à¤à¤ªà¥à¤¨ à¤à¤° à¤à¤à¤à¥à¤à¤ बनाता हà¥"},r3:{t:"सà¥à¤²à¥à¤¸ à¤à¤à¥à¤à¤",d:"à¤
वसरà¥à¤ à¤à¤¾ पà¥à¤°à¤¬à¤à¤§à¤¨"},r4:{t:"CFO à¤à¤à¥à¤à¤",d:"राà¤à¤¸à¥à¤µ à¤à¤° लाà¤à¤ªà¥à¤°à¤¦à¤¤à¤¾ à¤à¤¾ विशà¥à¤²à¥à¤·à¤£"},r5:{t:"à¤à¤¸à¥à¤à¤®à¤° सà¤à¥à¤¸à¥à¤¸ à¤à¤à¥à¤à¤",d:"à¤à¥à¤°à¤¾à¤¹à¤à¥à¤ à¤à¥ बनाठरà¤à¤¤à¤¾ हà¥"},r6:{t:"AI COO",d:"पà¥à¤°à¥ AI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸ à¤à¥ समनà¥à¤µà¤¿à¤¤ à¤à¤°à¤¤à¤¾ हà¥"}},final:"à¤à¤ª à¤à¥à¤à¤¬à¥à¤ नहà¥à¤ à¤à¤²à¤¾à¤¤à¥à¥¤ à¤à¤ª à¤à¤ पà¥à¤°à¥ AI à¤à¤à¤ªà¤¨à¥ à¤à¤²à¤¾à¤¤à¥ हà¥à¤à¥¤",cta:{primary:"à¤
पनॠAI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¥à¤",secondary:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ दà¥à¤à¥à¤"},links:{aria:"à¤à¤° à¤à¤¾à¤¨à¥à¤",workforce:"AI वरà¥à¤à¤«à¤¼à¥à¤°à¥à¤¸",marketplace:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ मारà¥à¤à¥à¤à¤ªà¥à¤²à¥à¤¸",pricing:"मà¥à¤²à¥à¤¯ निरà¥à¤§à¤¾à¤°à¤£",docs:"दसà¥à¤¤à¤¾à¤µà¥à¤à¤¼"}},ar:{eyebrow:"Ø§ÙØªÙ
ÙØ¶Ø¹",title:"ÙÙØ© عÙ
Ù AI Ù
ØªÙØ§Ù
ÙØ© Ù
بÙÙØ© ÙØ´Ø±ÙتÙ.",subtitle:"ØªÙØ¤ØªÙ
ت Salesforce ÙMicrosoft Copilot ÙServiceNow Ø³ÙØ± عÙ
Ù Ù
ØØ¯Ø¯Ø§Ù. Ø£Ù
ا Infernex ÙØªÙشر Ù
ÙØ¸ÙÙ AI ذاتÙÙÙ ÙØªØ¹Ø§ÙÙÙ٠عبر Ù
ؤسست٠باÙÙØ§Ù
Ù.",table:{feature:"اÙÙ
Ø¹ÙØ§Ø±",sf:"Salesforce",ms:"Microsoft Copilot",sn:"ServiceNow AI",ix:"Infernex AI"},rows:{product:{h:"اÙÙ
ÙØªØ¬ Ø§ÙØ£Ø³Ø§Ø³Ù",sf:"CRM + Ø¨ÙØ§Ùات Ø§ÙØ¹Ù
ÙØ§Ø¡",ms:"ØØ²Ù
Ø© Ø¥ÙØªØ§Ø¬ÙØ© + ÙØ¸Ø§Ù
تشغÙÙ Ù
ؤسسÙ",sn:"ITSM + Ø³ÙØ± عÙ
Ù Ù
ؤسسÙ",ix:"ÙÙØ© عÙ
Ù AI Ø°Ø§ØªÙØ© Ù
تعددة Ø§ÙØ£ÙساÙ
"},users:{h:"اÙÙ
ستخدÙ
ÙÙ Ø§ÙØ£Ø³Ø§Ø³ÙÙÙ",sf:"ÙØ±Ù اÙÙ
Ø¨ÙØ¹Ø§Øª",ms:"Ù
ÙØ¸ÙÙ Microsoft 365",sn:"ÙØ±Ù تÙÙÙØ© اÙÙ
عÙÙÙ
ات ÙØ§ÙعÙ
ÙÙØ§Øª",ix:"Ø§ÙØ´Ø±ÙØ© بأÙÙ
ÙÙØ§ + Ø§ÙØ¥Ø¯Ø§Ø±Ø© Ø§ÙØªÙÙÙØ°ÙØ©"},agents:{h:"ÙÙÙØ§Ø¡ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",sf:"ÙÙÙØ§Ø¡ Ù
تخصصÙ٠داخ٠Ù
ÙØ¸ÙÙ
Ø© Salesforce",ms:"Ù
ساعدÙÙ Ù
تصÙÙÙ Ø¨Ø¨ÙØ¦Ø© Microsoft",sn:"ÙÙÙØ§Ø¡ أتÙ
تة Ø³ÙØ± Ø§ÙØ¹Ù
Ù",ix:"Ù
ÙØ¸ÙÙ AI Ù
تخصصÙÙ ÙØ¹Ù
ÙÙÙ Ù
عاÙ"},dep:{h:"Ø§ÙØ§Ø¹ØªÙ
اد",sf:"اعتÙ
اد ÙØ¨Ùر عÙÙ Ù
ÙØ¸ÙÙ
Ø© Salesforce",ms:"اعتÙ
اد ÙØ¨Ùر عÙÙ Ù
ÙØ¸ÙÙ
Ø© Microsoft",sn:"اعتÙ
اد ÙØ¨Ùر عÙÙ Ù
ÙØ¸ÙÙ
Ø© ServiceNow",ix:"ÙØ¹Ù
٠باستÙÙØ§ÙÙØ© Ù
ع Ø£Ø¯ÙØ§ØªÙ Ø§ÙØØ§ÙÙØ©"},models:{h:"ÙÙ
اذج AI",sf:"Ù
ÙØ¸ÙÙ
ØªÙØ§ Ø¨Ø§ÙØ£Ø³Ø§Ø³",ms:"OpenAI + بÙÙØ© Microsoft",sn:"ÙÙ
اذج Ù
دÙ
جة",ix:"ØªÙØ¬Ù٠ذÙÙ Ù
تعدد اÙÙ LLM ÙÙØ§Ø²Ù بÙÙ Ø§ÙØªÙÙÙØ© ÙØ§Ùأداء"},deploy:{h:"اÙÙØ´Ø±",sf:"ÙØ´Ø± Ù
ؤسس٠طÙÙÙ",ms:"ÙØ´Ø± ÙØ§Ø¦Ù
عÙÙ Ø§ÙØªØ±Ø§Ø®Ùص",
811sn:"تÙÙÙØ° Ù
ؤسس٠ثÙÙÙ",ix:"ÙØ´Ø± SaaS Ø³Ø±ÙØ¹"},size:{h:"ØØ¬Ù
Ø§ÙØ´Ø±ÙØ©",sf:"ÙØ±ÙÙØ² عÙ٠اÙÙ
ؤسسات",ms:"Ù
٠اÙÙ
ØªÙØ³Ø·Ø© Ø¥Ù٠اÙÙØ¨Ùرة",sn:"Ø§ÙØ´Ø±Ùات اÙÙØ¨Ùرة",ix:"Ù
Ù Ø§ÙØ´Ø±Ùات اÙÙØ§Ø´Ø¦Ø© ÙØ§ÙØµØºÙØ±Ø© Ø¥ÙÙ Ø§ÙØ´Ø±Ùات Ø§ÙØ¹Ø§ÙÙ
ÙØ©"},market:{h:"Ø§ÙØ³ÙÙ",sf:"Ù
ÙØ¸ÙÙ
Ø© Salesforce",ms:"Ù
ÙØ¸ÙÙ
Ø© AppSource",sn:"Ù
تجر ServiceNow",ix:"سÙ٠عاÙÙ
Ù ÙÙ
ÙØ¸ÙÙ AI"}},os:{eyebrow:"اÙÙØ§Ø±Ù",title:"ÙØ¸Ø§Ù
تشغÙÙ ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",body:"ÙØ§ ØªØØªØ§Ø¬ Ø§ÙØ´Ø±Ùات Ø§ÙØµØºÙرة ÙÙØ§ اÙÙØ¨Ùرة Ø¥ÙÙ Ø±ÙØ¨Ùت Ù
ØØ§Ø¯Ø«Ø© آخر. ب٠إÙÙ ÙÙØ© عÙ
٠رÙÙ
ÙØ© ÙØ§Ù
ÙØ© Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ."},flow:{eyebrow:"Ù
ثا٠عÙÙ ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",title:"ÙØ±ÙÙ ÙÙ
٠اÙÙ
Ø¨ÙØ¹Ø§Øª",subtitle:"عدة Ù
ÙØ¸ÙÙ AI Ù
تصÙÙ٠عبر ÙØ¸Ø§Ù
تشغÙÙ Infernex AI Workforce.",orchestrator:"اÙÙ
ÙØ³ÙÙ",r1:{t:"SDR Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",d:"ÙØ¬Ø¯ اÙÙ
ØØªÙ
ÙÙÙ ÙÙØ¤ÙÙÙÙ
"},r2:{t:"ÙÙÙÙ Ø§ÙØªØ³ÙÙÙ",d:"ÙÙØ´Ø¦ Ø§ÙØÙ
ÙØ§Øª ÙØ§ÙÙ
ØØªÙÙ"},r3:{t:"ÙÙÙ٠اÙÙ
Ø¨ÙØ¹Ø§Øª",d:"ÙØ¯Ùر اÙÙØ±Øµ"},r4:{t:"ÙÙÙÙ CFO",d:"ÙØÙÙÙ Ø§ÙØ¥Ùرادات ÙØ§ÙربØÙØ©"},r5:{t:"ÙÙÙÙ ÙØ¬Ø§Ø Ø§ÙØ¹Ù
ÙØ§Ø¡",d:"ÙØØªÙØ¸ Ø¨Ø§ÙØ¹Ù
ÙØ§Ø¡"},r6:{t:"COO Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",d:"ÙÙØ³ÙÙ ÙÙØ© Ø§ÙØ¹Ù
٠باÙÙØ§Ù
Ù"}},final:"Ø£ÙØª ÙØ§ ØªØ¯ÙØ± Ø±ÙØ¨Ùت Ù
ØØ§Ø¯Ø«Ø©Ø Ø¨Ù ØªØ¯ÙØ± Ø´Ø±ÙØ© Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù ÙØ§Ù
ÙØ©.",cta:{primary:"Ø§ÙØ´Ø± ÙÙØ© عÙ
ÙÙ Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",secondary:"Ø§Ø³ØªÙØ´Ù Ù
ÙØ¸ÙÙ AI"},links:{aria:"اعر٠اÙÙ
Ø²ÙØ¯",workforce:"ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",marketplace:"سÙÙ Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",pricing:"Ø§ÙØªØ³Ø¹Ùر",docs:"اÙÙØ«Ø§Ø¦Ù"}}};for(const[e,t]of Object.entries(em)){const a=Ve[e];if(!a)continue;const o=a.wh??(a.wh={});o.diff=t}const tm={en:{r7:{t:"SEO Specialist",d:"Technical and generative SEO â every week"},r8:{t:"Competitor Intelligence",d:"24/7 monitoring of competitor moves"},r9:{t:"Data Analyst",d:"SQL, dashboards and insight memos"},r10:{t:"Compliance Officer",d:"Continuous SOC2, GDPR and ISO monitoring"},r11:{t:"Recruiter",d:"Sourcing, screening and coordination"},r12:{t:"Meeting Intelligence",d:"Every meeting transcribed and actioned"},viewAgent:"View agent",hireAll:"Explore all AI employees",countAgents:"{{n}} AI employees available",countConnectors:"{{n}} connectors available"},fr:{r7:{t:"Spécialiste SEO",d:"SEO technique et génératif â chaque semaine"},r8:{t:"Veille concurrentielle",d:"Surveillance 24/7 des concurrents"},r9:{t:"Analyste Data",d:"SQL, tableaux de bord et notes d'insight"},r10:{t:"Responsable Conformité",d:"Suivi continu SOC2, RGPD et ISO"},r11:{t:"Recruteur",d:"Sourcing, tri et coordination des entretiens"},r12:{t:"Meeting Intelligence",d:"Chaque réunion transcrite et actionnée"},viewAgent:"Voir l'agent",hireAll:"Explorer tous les employés IA",countAgents:"{{n}} employés IA disponibles",countConnectors:"{{n}} connecteurs disponibles"},es:{r7:{t:"Especialista SEO",d:"SEO técnico y generativo â cada semana"},r8:{t:"Inteligencia competitiva",d:"Monitoreo 24/7 de la competencia"},r9:{t:"Analista de Datos",d:"SQL, dashboards y memos de insights"},r10:{t:"Oficial de Cumplimiento",d:"Monitoreo continuo SOC2, GDPR e ISO"},r11:{t:"Reclutador",d:"Búsqueda, filtrado y coordinación"},r12:{t:"Meeting Intelligence",d:"Cada reunión transcrita y accionada"},viewAgent:"Ver agente",hireAll:"Explorar todos los empleados IA",countAgents:"{{n}} empleados IA disponibles",countConnectors:"{{n}} conectores disponibles"},de:{r7:{t:"SEO-Spezialist",d:"Technisches und generatives SEO â jede Woche"},r8:{t:"Wettbewerbs-Intelligenz",d:"24/7-Ãberwachung der Konkurrenz"},r9:{t:"Datenanalyst",d:"SQL, Dashboards und Insight-Memos"},r10:{t:"Compliance-Beauftragter",d:"Kontinuierliches SOC2-, DSGVO- und ISO-Monitoring"},r11:{t:"Recruiter",d:"Sourcing, Screening und Koordination"},r12:{t:"Meeting Intelligence",d:"Jedes Meeting transkribiert und umgesetzt"},viewAgent:"Agent ansehen",hireAll:"Alle KI-Mitarbeiter entdecken",countAgents:"{{n}} KI-Mitarbeiter verfügbar",countConnectors:"{{n}} Konnektoren verfügbar"},pt:{r7:{t:"Especialista SEO",d:"SEO técnico e generativo â toda semana"},r8:{t:"Inteligência de Concorrentes",d:"Monitoramento 24/7 dos concorrentes"},r9:{t:"Analista de Dados",d:"SQL, dashboards e memos de insights"},r10:{t:"Oficial de Conformidade",d:"Monitoramento contÃnuo SOC2, GDPR e ISO"}
811,r11:{t:"Recrutador",d:"Sourcing, triagem e coordenação"},r12:{t:"Meeting Intelligence",d:"Toda reunião transcrita e acionada"},viewAgent:"Ver agente",hireAll:"Explorar todos os funcionários IA",countAgents:"{{n}} funcionários IA disponÃveis",countConnectors:"{{n}} conectores disponÃveis"},zh:{r7:{t:"SEO ä¸å",d:"æ¯å¨ææ¯ä¸çæå¼ SEO"},r8:{t:"ç«äºæ
æ¥",d:"å
¨å¤©åçæ§ç«äºå¯¹æå¨æ"},r9:{t:"æ°æ®åæå¸",d:"SQLã仪表æ¿ä¸æ´å¯å¤å¿"},r10:{t:"åè§å®",d:"æç» SOC2ãGDPR ä¸ ISO çæ§"},r11:{t:"æèå®",d:"寻æºãçéä¸é¢è¯åè°"},r12:{t:"ä¼è®®æºè½",d:"æ¯æ¬¡ä¼è®®èªå¨è½¬åå¹¶æ§è¡"},viewAgent:"æ¥ç该åå·¥",hireAll:"æµè§ææ AI åå·¥",countAgents:"å¯ç¨ AI åå·¥ {{n}} å",countConnectors:"å¯ç¨è¿æ¥å¨ {{n}} 个"},ja:{r7:{t:"SEOã¹ãã·ã£ãªã¹ã",d:"æè¡ã»çæSEOãæ¯é±å®è¡"},r8:{t:"ç«¶åã¤ã³ããªã¸ã§ã³ã¹",d:"ç«¶åã24æé365æ¥ç£è¦"},r9:{t:"ãã¼ã¿ã¢ããªã¹ã",d:"SQLã»ããã·ã¥ãã¼ãã»ã¤ã³ãµã¤ã"},r10:{t:"ã³ã³ãã©ã¤ã¢ã³ã¹è²¬ä»»è
",d:"SOC2ã»GDPRã»ISOãç¶ç¶ç£è¦"},r11:{t:"ãªã¯ã«ã¼ã¿ã¼",d:"ã½ã¼ã·ã³ã°ã»ã¹ã¯ãªã¼ãã³ã°ã»èª¿æ´"},r12:{t:"ãã¼ãã£ã³ã°ã»ã¤ã³ããªã¸ã§ã³ã¹",d:"å
¨ä¼è°ãæ¸ãèµ·ããã¢ã¯ã·ã§ã³å"},viewAgent:"ã¨ã¼ã¸ã§ã³ããè¦ã",hireAll:"ãã¹ã¦ã®AI徿¥å¡ãè¦ã",countAgents:"å©ç¨å¯è½ãªAI徿¥å¡ {{n}} å",countConnectors:"å©ç¨å¯è½ãªã³ãã¯ã¿ {{n}} å"},hi:{r7:{t:"SEO विशà¥à¤·à¤à¥à¤",d:"सापà¥à¤¤à¤¾à¤¹à¤¿à¤ तà¤à¤¨à¥à¤à¥ à¤à¤° à¤à¤¨à¤°à¥à¤à¤¿à¤µ SEO"},r8:{t:"पà¥à¤°à¤¤à¤¿à¤¸à¥à¤ªà¤°à¥à¤§à¥ à¤à¤à¤à¥à¤²à¤¿à¤à¥à¤à¤¸",d:"24/7 पà¥à¤°à¤¤à¤¿à¤¸à¥à¤ªà¤°à¥à¤§à¤¿à¤¯à¥à¤ à¤à¥ निà¤à¤°à¤¾à¤¨à¥"},r9:{t:"डà¥à¤à¤¾ विशà¥à¤²à¥à¤·à¤",d:"SQL, डà¥à¤¶à¤¬à¥à¤°à¥à¤¡ à¤à¤° à¤à¤¨à¤¸à¤¾à¤à¤ मà¥à¤®à¥"},r10:{t:"à¤
नà¥à¤ªà¤¾à¤²à¤¨ à¤
धिà¤à¤¾à¤°à¥",d:"निरà¤à¤¤à¤° SOC2, GDPR à¤à¤° ISO निà¤à¤°à¤¾à¤¨à¥"},r11:{t:"रिà¤à¥à¤°à¥à¤à¤°",d:"सà¥à¤°à¥à¤¸à¤¿à¤à¤, सà¥à¤à¥à¤°à¥à¤¨à¤¿à¤à¤ à¤à¤° समनà¥à¤µà¤¯"},r12:{t:"मà¥à¤à¤¿à¤à¤ à¤à¤à¤à¥à¤²à¤¿à¤à¥à¤à¤¸",d:"हर मà¥à¤à¤¿à¤à¤ à¤à¥à¤°à¤¾à¤à¤¸à¤à¥à¤°à¤¾à¤à¤¬ à¤à¤° à¤à¤à¥à¤¶à¤¨"},viewAgent:"à¤à¤à¥à¤à¤ दà¥à¤à¥à¤",hireAll:"सà¤à¥ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ दà¥à¤à¥à¤",countAgents:"{{n}} AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¤ªà¤²à¤¬à¥à¤§",countConnectors:"{{n}} à¤à¤¨à¥à¤à¥à¤à¤° à¤à¤ªà¤²à¤¬à¥à¤§"},ar:{r7:{t:"أخصائ٠SEO",d:"SEO تÙÙÙ ÙØªÙÙÙØ¯Ù Ø£Ø³Ø¨ÙØ¹ÙÙØ§"},r8:{t:"استخبارات اÙÙ
ÙØ§ÙسÙÙ",d:"Ù
Ø±Ø§ÙØ¨Ø© اÙÙ
ÙØ§ÙسÙ٠عÙÙ Ù
دار Ø§ÙØ³Ø§Ø¹Ø©"},r9:{t:"Ù
ØÙÙ Ø¨ÙØ§Ùات",d:"SQL ÙÙÙØØ§Øª اÙÙÙØ§Ø¯Ø© ÙÙ
Ø°ÙØ±Ø§Øª Ø§ÙØ±Ø¤Ù"},r10:{t:"Ù
سؤÙÙ Ø§ÙØ§Ù
تثاÙ",d:"Ù
Ø±Ø§ÙØ¨Ø© Ù
ستÙ
رة ÙÙ SOC2 ÙGDPR ÙISO"},r11:{t:"Ù
سؤÙÙ Ø§ÙØªÙظÙÙ",d:"Ø¨ØØ« ÙÙØ±Ø² ÙØªÙسÙ٠اÙÙ
ÙØ§Ø¨Ùات"},r12:{t:"Ø°ÙØ§Ø¡ Ø§ÙØ§Ø¬ØªÙ
اعات",d:"ØªÙØ±Ùغ ÙØªÙÙÙØ° Ù٠اجتÙ
اع"},viewAgent:"عرض اÙÙ
ÙØ¸Ù",hireAll:"Ø§Ø³ØªÙØ´Ø§Ù ÙÙ Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",countAgents:"{{n}} Ù
ÙØ¸Ù Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù Ù
تاØ",countConnectors:"{{n}} Ù
ÙØµÙÙØ§ Ù
تاØÙا"}};for(const[e,t]of Object.entries(tm)){const a=Ve[e];if(!a)continue;const o=a.wh,r=o==null?void 0:o.diff,s=r==null?void 0:r.flow;s&&Object.assign(s,t)}const nm={en:{banner:{newBadge:"NEW",eyebrow:"Master AI Orchestrator",masterTitle:"Master AI Agent",title:"From an idea to an operational AI-powered company.",sub:"With Infernex AI Workforce, anyone can transform an idea into an operational company assisted by a complete AI workforce, from conception to execution.",ctaAuto:"Activate the master AI orchestrator",ctaManual:"Build my workforce manually"}},fr:{banner:{newBadge:"NOUVEAU",eyebrow:"Orchestrateur IA Maître",masterTitle:"Agent IA Maître",title:"D'une idée à une entreprise opérationnelle propulsée par l'IA.",sub:"Avec Infernex AI Workforce, quiconque peut transformer une idée en entreprise opérationnelle assistée par une force de travail IA complète, de la conception à l'exécution.",ctaAuto:"Activer l'orchestrateur",ctaManual:"Construire ma force de travail manuellement"}},es:{banner:{newBadge:"NUEVO",eyebrow:"Orquestador IA Maestro",masterTitle:"Agente IA Maestro",title:"De una idea a una empresa operativa impulsada por IA.",sub:"Con Infernex AI Workforce, cualquiera puede transformar una idea en una empresa operativa asistida por una fuerza laboral de IA completa, de la concepción a la ejecución.",ctaAuto:"Activar el orquestador",ctaManual:"Construir mi fuerza laboral manualmente"}},de:{banner:{newBadge:"NEU",eyebrow:"Master-KI-Orchestrator",masterTitle:"Master-KI-Agent",title:"Von einer Idee zu einem operativen KI-gesteuerten Unternehmen.",sub:"Mit Infernex AI Workforce kann jeder eine Idee in ein operatives Unternehmen verwandeln, das von einer vollständigen KI-Belegschaft unterstützt wird â von der Konzeption bis zur Ausführung.",ctaAuto:"Orchestrator aktivieren",ctaManual:"Meine KI-Belegschaft manuell aufbauen"}},pt:{banner:{newBadge:"NOVO",eyebrow:"Orquestrador IA Mestre",masterTitle:"Agente IA Mestre",title:"De uma ideia a uma empresa operacional impulsionada por IA.",sub:"Com a Infernex AI Workforce, qualquer pessoa pode transformar uma ideia em uma empresa operacional assistida por uma força de trabalho de IA completa, da concepção à execução.",ctaAuto:"Ativar o orquestrador",ctaManual:"Construir minha força de trabalho manualmente"}},zh:{banner:{newBadge:"æ°åè½",eyebrow:"AI æ»ææ¥",masterTitle:"AI æ»ä»£ç",title:"ä»ä¸ä¸ªæ³æ³å°ç± AI 驱å¨çè¿è¥åä¼ä¸ã",sub:"åå© Infernex AI Workforceï¼ä»»ä½äººé½å¯ä»¥å°ä¸ä¸ªæ³æ³è½¬å为ä¸ä¸ªç±å®æ´ AI åå·¥å¢éè¾
å©è¿è¥çæçä¼ä¸ï¼ä»ææå°æ§è¡ã",ctaAuto:"å¯å¨æ»ææ¥",ctaManual:"æå¨æå»ºæç AI åå·¥å¢é"}},ja:{banner:{newBadge:"æ°æ©è½",eyebrow:"ãã¹ã¿ã¼AIãªã¼ã±ã¹ãã¬ã¼ã¿ã¼",masterTitle:"ãã¹ã¿ã¼AIã¨ã¼ã¸ã§ã³ã",title:"ã¢ã¤ãã¢ãããAIã§éå¶ããã伿¥ã¸ã",sub:"Infernex AI Workforce ã使ãã°ã誰ããã¢ã¤ãã¢ããå®å
¨ãª AI ã¯ã¼ã¯ãã©ã¼ã¹ã«æ¯ããããéå¶ä¼æ¥ã«å¤é©ã§ãã¾ããä¼ç»ããå®è¡ã¾ã§ã",ctaAuto:"ãªã¼ã±ã¹ãã¬ã¼ã¿ã¼ãèµ·å",ctaManual:"æåã§ã¯ã¼ã¯ãã©ã¼ã¹ãæ§ç¯"}},hi:{banner:{newBadge:"नया",eyebrow:"मासà¥à¤à¤° AI à¤à¤°à¥à¤à¥à¤¸à¥à¤à¥à¤°à¥à¤à¤°",masterTitle:"मासà¥à¤à¤° AI à¤à¤à¥à¤à¤",title:"à¤à¤ विà¤à¤¾à¤° सॠà¤à¤ AI-सà¤à¤à¤¾à¤²à¤¿à¤¤ परिà¤à¤¾à¤²à¤¨ à¤à¤à¤ªà¤¨à¥ तà¤à¥¤",sub:"Infernex AI Workforce à¤à¥ साथ, à¤à¥à¤ à¤à¥ à¤à¤ विà¤à¤¾à¤° à¤à¥ पà¥à¤°à¥ AI à¤à¤¾à¤°à¥à¤¯à¤¬à¤² à¤à¥ सहायता सॠपरिà¤à¤¾à¤²à¤¨ à¤à¤à¤ªà¤¨à¥ मà¥à¤ बदल सà¤à¤¤à¤¾ हॠâ à¤
वधारणा सॠनिषà¥à¤ªà¤¾à¤¦à¤¨ तà¤à¥¤",ctaAuto:"à¤à¤°à¥à¤à¥à¤¸à¥à¤à¥à¤°à¥à¤à¤° सà¤à¥à¤°à¤¿à¤¯ à¤à¤°à¥à¤",ctaManual:"मà¥à¤°à¤¾ à¤à¤¾à¤°à¥à¤¯à¤¬à¤² सà¥à¤µà¤¯à¤ बनाà¤à¤"}},ar:{banner:{newBadge:"Ø¬Ø¯ÙØ¯",eyebrow:"Ù
ÙØ³ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù Ø§ÙØ±Ø¦ÙسÙ",masterTitle:"ÙÙÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù Ø§ÙØ±Ø¦ÙسÙ",title:"Ù
Ù ÙÙØ±Ø© Ø¥ÙÙ Ø´Ø±ÙØ© تشغÙÙÙØ© Ù
دعÙÙ
Ø© Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ.",sub:"Ù
ع Infernex AI WorkforceØ ÙÙ
ÙÙ ÙØ£Ù شخص تØÙÙÙ ÙÙØ±Ø© Ø¥ÙÙ Ø´Ø±ÙØ© تشغÙÙÙØ© Ù
دعÙÙ
Ø© بÙÙØ© عÙ
Ù AI ÙØ§Ù
ÙØ©Ø Ù
Ù Ø§ÙØªØµÙر Ø¥ÙÙ Ø§ÙØªÙÙÙØ°.",ctaAuto:"ØªÙØ¹Ù٠اÙÙ
ÙØ³ÙÙ",ctaManual:"Ø¨ÙØ§Ø¡ ÙÙØ© Ø§ÙØ¹Ù
Ù ÙØ¯ÙÙÙØ§"}}};for(const[e,t]of Object.entries(nm)){const a=Ve[e]??(Ve[e]={});a.orch=t}const Xt={en:{department:{workforce_command:"Workforce Command"}
811,employeeItems:{"ai-workforce-commander":{name:"AI Workforce Commander (AI COO)",tagline:"Virtual Chief Operating Officer orchestrating your entire AI workforce.",overview:"AI Workforce Commander is the central operational intelligence layer of Infernex. It monitors every deployed AI Employee, coordinates multi-agent workflows, detects bottlenecks, optimizes business processes and generates executive reports for the C-suite in real time.",businessValue:["One AI COO overseeing the whole AI workforce","Detects workflow bottlenecks across departments","Recommends automation opportunities weekly","Board-grade operational KPI reporting","24/7 alerting on operational anomalies"],workflows:["Company-wide workflow scan","Multi-agent orchestration plan","Bottleneck & idle-agent report","Executive KPI briefing","Automation opportunity backlog"],roi:"Saves $260k / year in ops leadership and unlocks $3.5M / year in productivity gains per year.",faq:[{q:"Does it replace a human COO?",a:"No â it augments operational leadership with 24/7 visibility, orchestration and reporting across every AI Employee you deploy."},{q:"Which AI Employees can it coordinate?",a:"Every Infernex AI Employee, plus third-party agents connected via MCP or the Infernex API."},{q:"Is the operational data isolated?",a:"Yes â tenant-isolated with AES-256 encryption and full audit trail."}]},"ai-knowledge-manager":{name:"AI Knowledge Manager",tagline:"Enterprise AI brain â every internal document, instantly answerable.",overview:"AI Knowledge Manager ingests company documents from Drive, SharePoint, Notion and Confluence to build a live internal knowledge base. Employees ask any question in their own language and get answers grounded in your latest official documentation, with outdated-content detection built in.",businessValue:["Instant answers on any internal policy or process","Auto-detected outdated documentation","Multilingual knowledge assistant for every team","Reduced onboarding time by 40-60%","Zero knowledge loss when employees leave"],workflows:["Corpus ingestion & indexing","Internal Q&A assistant","Documentation freshness audit","Knowledge gap detection","Auto-drafted policy updates"],roi:"Saves $140k / year in support & search and unlocks $900k / year in reclaimed productivity per year.",faq:[{q:"Where is the knowledge stored?",a:"In your tenant-isolated Infernex vector store â encrypted at rest, never used to train shared models."},{q:"How often is content refreshed?",a:"Continuous incremental sync, with a full re-index on demand."},{q:"Does it respect source permissions?",a:"Yes â ACLs from Drive, SharePoint, Notion and Confluence are preserved at query time."}]},"ai-revenue-intelligence":{name:"AI Revenue Intelligence Agent",tagline:"Full-stack revenue intelligence â pipeline, forecast, risk.",overview:"AI Revenue Intelligence continuously analyses your CRM, billing and product-usage data to forecast revenue, flag at-risk deals and recommend the next best sales action for every account.",businessValue:["Sales pipeline analysed in real time","AI-generated revenue forecast every week","Deal-risk & churn early-warning","Customer behaviour segmentation","Recommended next best action per account"],workflows:["Pipeline health scoring","Weekly revenue forecast","Lost-deal post-mortem","Customer behaviour cohort analysis","Next-best-action recommender"],roi:"Saves $180k / year in RevOps and unlocks $2.4M / year in recovered pipeline per year.",faq:[{q:"Which CRMs are supported?",a:"Salesforce, HubSpot, Pipedrive and Microsoft Dynamics 365 â via the Connector Marketplace."},{q:"How accurate is the forecast?",a:"Typically 92-97% MAPE within one quarter after connection, tuned on your own history."},{q:"Can it push actions to the CRM?",a:"Yes â with human approval workflows and full audit log."}]},"ai-process-automation-auditor":{name:"AI Process Automation Auditor",tagline:"Find every hidden automation opportunity inside your company.",overview:"AI Process Automation Auditor scans repetitive activity across your Google Workspace, Microsoft 365, chat, ERP and CRM to surface the highest-ROI automation opportunities, prioritized by estimated saved cost.",businessValue:["Continuous automation opportunity discovery","Repetitive-task heatmap by team","Ranked automation backlog with cost savings","Zero-effort ROI estimation per workflow","Enterprise-wide efficiency benchmark"],workflows:["Activity mining & task clustering","Inefficient workflow diagnosis","Automation backlog & ROI ranking","Optimization report for the C-suite","Handover to AI Employees for execution"],roi:"Saves $160k / year in o
811ps analysis and unlocks $2M+ / year in eliminated repetitive work per year.",faq:[{q:"Does it need to read our documents?",a:"It reads workflow metadata by default. Full content access is opt-in per source."},{q:"Can it plug into any ERP or CRM?",a:"Yes â SAP, Oracle, NetSuite, Salesforce, HubSpot, Dynamics 365 and Postgres/MySQL/BigQuery/Snowflake."},{q:"How fast until we see results?",a:"First ranked automation backlog delivered within 7 days of connecting your sources."}]},"ai-employee-manager":{name:"AI Employee Manager",tagline:"Manage the entire AI workforce ecosystem â performance, cost and permissions.",overview:"AI Employee Manager tracks every AI Employee's performance, token consumption and cost. It recommends model optimizations, flags failed tasks, manages permissions and produces a full AI workforce report for the CFO and CTO.",businessValue:["Live performance monitoring for every AI Employee","Token & cost tracking down to the task","Automatic model routing recommendations","Failed-task detection & auto-retry policies","AI workforce compliance & permissions"],workflows:["AI Employee performance dashboard","Token consumption & cost analysis","Model optimization recommendation","Failed-task queue triage","Permissions & audit report"],roi:"Saves $120k / year in AI ops and unlocks $1.5M / year in inference cost savings per year.",faq:[{q:"Which AI providers are tracked?",a:"Infernex API, OpenAI, Anthropic, Google Gemini, Mistral, Azure AI and any custom endpoint."},{q:"Can it enforce spend limits?",a:"Yes â per employee, per team, per model, with hard caps and alerting."},{q:"Is it required for other AI Employees?",a:"Recommended for anyone running 3+ AI Employees in production."}]}}},fr:{department:{workforce_command:"Command Center IA"},employeeItems:{"ai-workforce-commander":{name:"Commandant AI Workforce (COO IA)",tagline:"Directeur des opérations virtuel qui orchestre toute votre force de travail IA.",overview:"AI Workforce Commander est la couche centrale dâintelligence opérationnelle dâInfernex. Il surveille chaque employé IA déployé, coordonne les workflows multi-agents, détecte les goulots dâétranglement, optimise les processus métier et génère en temps réel des rapports exécutifs pour le comité de direction.",businessValue:["Un COO IA qui supervise toute la force de travail IA","Détecte les goulots dâétranglement entre départements","Recommande chaque semaine des opportunités dâautomatisation","Reporting KPI opérationnel de niveau comité de direction","Alertes 24/7 sur les anomalies opérationnelles"],workflows:["Scan des workflows de toute lâentreprise","Plan dâorchestration multi-agents","Rapport goulots dâétranglement et agents inactifs","Briefing KPI exécutif","Backlog dâopportunités dâautomatisation"],roi:"Ãconomise 260 k$ / an en leadership opérationnel et débloque 3,5 M$ / an de gains de productivité.",faq:[{q:"Remplace-t-il un COO humain ?",a:"Non â il renforce la direction opérationnelle avec une visibilité, une orchestration et un reporting 24/7 sur chaque employé IA déployé."},{q:"Quels employés IA peut-il coordonner ?",a:"Tous les employés IA Infernex, ainsi que les agents tiers connectés via MCP ou lâAPI Infernex."},{q:"Les données opérationnelles sont-elles isolées ?",a:"Oui â isolation par tenant, chiffrement AES-256 et piste dâaudit complète."}]},"ai-knowledge-manager":{name:"Gestionnaire de connaissance IA",tagline:"Cerveau IA dâentreprise â chaque document interne devient interrogeable instantanément.",overview:"AI Knowledge Manager ingère les documents Drive, SharePoint, Notion et Confluence pour créer une base de connaissance interne vivante. Les employés posent leurs questions dans leur langue et obtiennent des réponses ancrées dans la documentation officielle la plus récente, avec détection intégrée du contenu obsolète.",businessValue:["Réponses instantanées sur toute politique ou processus interne","Documentation obsolète détectée automatiquement","Assistant de connaissance multilingue pour chaque équipe","Temps dâonboarding réduit de 40 à 60 %","Aucune perte de connaissance lors des départs"],workflows:["Ingestion et indexation du corpus","Assistant Q/R interne","Audit de fraîcheur documentaire","Détection des lacunes de connaissance","Mise
811s à jour de politiques pré-rédigées"],roi:"Ãconomise 140 k$ / an en support et recherche interne et débloque 900 k$ / an de productivité récupérée.",faq:[{q:"Où la connaissance est-elle stockée ?",a:"Dans votre vector store Infernex isolé par tenant â chiffré au repos et jamais utilisé pour entraîner des modèles partagés."},{q:"à quelle fréquence le contenu est-il rafraîchi ?",a:"Synchronisation incrémentale continue, avec réindexation complète à la demande."},{q:"Respecte-t-il les permissions source ?",a:"Oui â les ACL de Drive, SharePoint, Notion et Confluence sont préservées lors des requêtes."}]},"ai-revenue-intelligence":{name:"Agent IA Revenue Intelligence",tagline:"Intelligence revenue complète â pipeline, prévisions, risques.",overview:"AI Revenue Intelligence analyse en continu votre CRM, votre facturation et vos données dâusage produit pour prévoir le chiffre dâaffaires, signaler les deals à risque et recommander la meilleure prochaine action commerciale pour chaque compte.",businessValue:["Pipeline commercial analysé en temps réel","Prévision de revenus générée par IA chaque semaine","Alerte précoce sur risques de deal et churn","Segmentation des comportements clients","Meilleure prochaine action recommandée par compte"],workflows:["Scoring santé du pipeline","Prévision revenue hebdomadaire","Post-mortem des deals perdus","Analyse cohortes comportement client","Recommandation de next-best-action"],roi:"Ãconomise 180 k$ / an en RevOps et débloque 2,4 M$ / an de pipeline récupéré.",faq:[{q:"Quels CRM sont supportés ?",a:"Salesforce, HubSpot, Pipedrive et Microsoft Dynamics 365 â via la marketplace de connecteurs."},{q:"Quelle est la précision des prévisions ?",a:"Généralement 92 à 97 % MAPE sur un trimestre après connexion, ajusté sur votre historique."},{q:"Peut-il pousser des actions dans le CRM ?",a:"Oui â avec workflows dâapprobation humaine et journal dâaudit complet."}]},"ai-process-automation-auditor":{name:"Auditeur IA dâautomatisation des processus",tagline:"Trouve chaque opportunité dâautomatisation cachée dans votre entreprise.",overview:"AI Process Automation Auditor scanne lâactivité répétitive dans Google Workspace, Microsoft 365, chat, ERP et CRM pour faire émerger les opportunités dâautomatisation au meilleur ROI, priorisées par économies estimées.",businessValue:["Découverte continue dâopportunités dâautomatisation","Heatmap des tâches répétitives par équipe","Backlog dâautomatisation classé par économies","Estimation ROI sans effort pour chaque workflow","Benchmark dâefficacité à lâéchelle de lâentreprise"],workflows:["Mining dâactivité et clustering de tâches","Diagnostic des workflows inefficaces","Backlog dâautomatisation et classement ROI","Rapport dâoptimisation pour le comité de direction","Passage aux employés IA pour exécution"],roi:"Ãconomise 160 k$ / an en analyse opérationnelle et débloque 2 M$+ / an de travail répétitif éliminé.",faq:[{q:"Doit-il lire nos documents ?",a:"Il lit par défaut les métadonnées de workflow. Lâaccès au contenu complet est optionnel par source."},{q:"Peut-il se connecter à un ERP ou CRM ?",a:"Oui â SAP, Oracle, NetSuite, Salesforce, HubSpot, Dynamics 365 et Postgres/MySQL/BigQuery/Snowflake."},{q:"En combien de temps voit-on les résultats ?",a:"Premier backlog dâautomatisation priorisé livré sous 7 jours après connexion des sources."}]},"ai-employee-manager":{name:"Manager dâemployés IA",tagline:"Pilote tout lâécosystème AI Workforce â performance, coûts et permissions.",overview:"AI Employee Manager suit la performance, la consommation de tokens et les coûts de chaque employé IA. Il recommande des optimisations de modèles, signale les tâches échouées, gère les permissions et produit un rapport complet AI Workforce pour le CFO et le CTO.",businessValue:["Monitoring live de chaque employé IA","Suivi tokens et coûts jusquâà la tâche","Recommandations automatiques de routage modèle","Détection des tâches échouées et politiques dâauto-retry","Conformité et permissions de la force de travail IA"],workflows:["Dashboard performance des employés IA","Analyse consommation tokens et coûts","Recommandation dâoptimisation modèle","Triage file des tâches échouées","Rapport permissions et audit"],roi:"Ãconomise 120 k$ / an en AI Ops et débloque 1,5 M$ / an dâéconomies sur les coûts dâinférence.",faq:[{q:"Quels fournisseurs IA sont suivis ?",a:"API Infernex, OpenAI, Anthropic, Google Gemini, Mistral, Azure AI et tout endpoint personnalisé."},{q:"Peut-il imposer des limites de dépense ?",a:"Oui â par employé, équipe et modèle, avec hard caps et alertes."},{q:"Est-il requis pour les autres employés IA ?",a:"Recommandé dès que vous exploitez 3+ employés IA en production."}]}}},es:{department:{workforce_command:"Centro de mando IA"}
811,employeeItems:{"ai-workforce-commander":{name:"Comandante AI Workforce (COO IA)",tagline:"Director de operaciones virtual que orquesta toda tu fuerza laboral de IA.",overview:"AI Workforce Commander es la capa central de inteligencia operativa de Infernex. Supervisa cada empleado IA desplegado, coordina flujos multiagente, detecta cuellos de botella, optimiza procesos de negocio y genera informes ejecutivos para la dirección en tiempo real.",businessValue:["Un COO IA supervisando toda la fuerza laboral","Detecta cuellos de botella entre departamentos","Recomienda oportunidades de automatización cada semana","Reporting KPI operativo de nivel consejo","Alertas 24/7 sobre anomalÃas operativas"],workflows:["Escaneo de flujos de toda la empresa","Plan de orquestación multiagente","Informe de cuellos de botella y agentes inactivos","Briefing ejecutivo de KPI","Backlog de oportunidades de automatización"],roi:"Ahorra 260 k$ / año en liderazgo operativo y desbloquea 3,5 M$ / año en productividad.",faq:[{q:"¿Sustituye a un COO humano?",a:"No â aumenta el liderazgo operativo con visibilidad, orquestación y reporting 24/7 sobre cada empleado IA desplegado."},{q:"¿Qué empleados IA puede coordinar?",a:"Todos los empleados IA de Infernex, además de agentes externos conectados por MCP o la API de Infernex."},{q:"¿Los datos operativos están aislados?",a:"Sà â aislamiento por tenant, cifrado AES-256 y auditorÃa completa."}]},"ai-knowledge-manager":{name:"Gestor de conocimiento IA",tagline:"Cerebro IA empresarial â cada documento interno, respondible al instante.",overview:"AI Knowledge Manager ingiere documentos de Drive, SharePoint, Notion y Confluence para construir una base de conocimiento interna viva con respuestas en el idioma de cada empleado.",businessValue:["Respuestas instantáneas sobre polÃticas y procesos","Documentación obsoleta detectada automáticamente","Asistente multilingüe para cada equipo","Onboarding reducido un 40-60%","Sin pérdida de conocimiento al salir empleados"],workflows:["Ingesta e indexación del corpus","Asistente interno de preguntas","AuditorÃa de frescura documental","Detección de lagunas de conocimiento","Actualizaciones de polÃticas pre-redactadas"],roi:"Ahorra 140 k$ / año en soporte y búsqueda y desbloquea 900 k$ / año en productividad recuperada.",faq:[{q:"¿Dónde se almacena el conocimiento?",a:"En tu vector store Infernex aislado por tenant, cifrado en reposo y nunca usado para entrenar modelos compartidos."},{q:"¿Cada cuánto se actualiza?",a:"Sincronización incremental continua, con reindexación completa bajo demanda."},{q:"¿Respeta permisos de origen?",a:"Sà â conserva ACL de Drive, SharePoint, Notion y Confluence."}]},"ai-revenue-intelligence":{name:"Agente IA de Revenue Intelligence",tagline:"Inteligencia de ingresos completa â pipeline, previsión y riesgo.",overview:"AI Revenue Intelligence analiza CRM, facturación y uso de producto para prever ingresos, detectar deals en riesgo y recomendar la siguiente mejor acción comercial.",businessValue:["Pipeline analizado en tiempo real","Previsión de ingresos semanal por IA","Alerta temprana de riesgo y churn","Segmentación de comportamiento cliente","Siguiente mejor acción por cuenta"],workflows:["Scoring de salud del pipeline","Previsión semanal de ingresos","Post-mortem de deals perdidos","Análisis de cohortes","Recomendador de siguiente acción"],roi:"Ahorra 180 k$ / año en RevOps y desbloquea 2,4 M$ / año en pipeline recuperado.",faq:[{q:"¿Qué CRM soporta?",a:"Salesforce, HubSpot, Pipedrive y Microsoft Dynamics 365 vÃa marketplace de conectores."},{q:"¿Qué tan precisa es la previsión?",a:"Normalmente 92-97% MAPE tras un trimestre conectado y ajustado a tu histórico."},{q:"¿Puede ejecutar acciones en el CRM?",a:"SÃ, con aprobación humana y auditorÃa completa."}]},"ai-process-automation-auditor":{name:"Auditor IA de automatización de procesos",tagline:"Encuentra cada oportunidad oculta de automatización en tu empresa.",overview:"AI Process Automation Auditor analiza actividad repetitiva en workspace, chat, ERP y CRM para priorizar automatizaciones por ROI estimado.",businessValue:["Descubrimiento continuo de automatizaciones","Mapa de calor de tareas repetitivas","Backlog priorizado por ahorro","ROI estimado por workflow","Benchmark de eficiencia empresarial"],workflows:["MinerÃa de actividad y clustering","Diagnóstico de workflows ineficientes","Backlog y ranking ROI","Informe para la dirección","Traspaso a empleados IA"],roi:"Ahorra 160 k$ / año en análisis operativo y desbloquea 2 M$+ / año eliminando trabajo repetitivo.",faq:[{q:"¿Necesita leer documentos?",a:"Lee metadatos por defecto; el contenido completo es opcional por fuente."},{q:"¿Conecta con ERP o CRM?",a:"Sà â SAP, Oracle, NetSuite, Salesforce, HubSpot, Dynamics 365 y base
811s de datos."},{q:"¿Cuándo hay resultados?",a:"Primer backlog priorizado en 7 dÃas tras conectar fuentes."}]},"ai-employee-manager":{name:"Manager de empleados IA",tagline:"Gestiona todo el ecosistema AI Workforce â rendimiento, coste y permisos.",overview:"AI Employee Manager sigue rendimiento, tokens y coste de cada empleado IA, recomienda optimizaciones, detecta fallos, gestiona permisos y genera informes para CFO y CTO.",businessValue:["Monitorización en vivo de cada empleado IA","Tokens y costes hasta nivel tarea","Recomendaciones de enrutamiento de modelo","Detección de fallos y auto-retry","Cumplimiento y permisos AI Workforce"],workflows:["Dashboard de rendimiento","Análisis de tokens y costes","Optimización de modelo","Triage de tareas fallidas","Informe de permisos y auditorÃa"],roi:"Ahorra 120 k$ / año en AI Ops y desbloquea 1,5 M$ / año en ahorro de inferencia.",faq:[{q:"¿Qué proveedores IA rastrea?",a:"API Infernex, OpenAI, Anthropic, Google Gemini, Mistral, Azure AI y endpoints personalizados."},{q:"¿Puede imponer lÃmites de gasto?",a:"Sà â por empleado, equipo y modelo, con hard caps y alertas."},{q:"¿Es obligatorio?",a:"Recomendado para 3+ empleados IA en producción."}]}}},de:{department:{workforce_command:"KI-Kommandozentrale"},employeeItems:{"ai-workforce-commander":{name:"AI Workforce Commander (KI-COO)",tagline:"Virtueller COO, der Ihre gesamte KI-Belegschaft orchestriert.",overview:"AI Workforce Commander ist die zentrale operative Intelligenzschicht von Infernex. Er überwacht jeden KI-Mitarbeiter, koordiniert Multi-Agenten-Workflows, erkennt Engpässe, optimiert Prozesse und erstellt Executive Reports in Echtzeit.",businessValue:["Ein KI-COO für die gesamte KI-Belegschaft","Erkennt Engpässe über Abteilungen hinweg","Empfiehlt wöchentlich Automatisierungen","KPI-Reporting auf Vorstandsniveau","24/7 Alerts bei operativen Anomalien"],workflows:["Unternehmensweiter Workflow-Scan","Multi-Agenten-Orchestrierungsplan","Engpass- und Idle-Agent-Report","Executive KPI Briefing","Automatisierungs-Backlog"],roi:"Spart 260 Tsd. $ / Jahr in operativer Führung und erschlieÃt 3,5 Mio. $ / Jahr Produktivitätsgewinn.",faq:[{q:"Ersetzt er einen menschlichen COO?",a:"Nein â er erweitert die operative Führung mit 24/7 Sichtbarkeit, Orchestrierung und Reporting."},{q:"Welche KI-Mitarbeiter kann er koordinieren?",a:"Alle Infernex KI-Mitarbeiter plus Drittanbieter-Agenten via MCP oder Infernex API."},{q:"Sind operative Daten isoliert?",a:"Ja â Tenant-Isolation, AES-256 Verschlüsselung und vollständiger Audit-Trail."}]},"ai-knowledge-manager":{name:"KI-Wissensmanager",tagline:"Enterprise-KI-Gehirn â jedes interne Dokument sofort beantwortbar.",overview:"AI Knowledge Manager verarbeitet Dokumente aus Drive, SharePoint, Notion und Confluence zu einer lebenden Wissensbasis mit Antworten in der Sprache jedes Mitarbeiters.",businessValue:["Sofortantworten zu Richtlinien und Prozessen","Automatische Erkennung veralteter Dokumente","Mehrsprachiger Wissensassistent","Onboarding 40-60% schneller","Kein Wissensverlust bei Abgängen"],workflows:["Korpus-Ingestion und Indexierung","Interner Q&A-Assistent","Dokumenten-Frischeaudit","Wissenslücken-Erkennung","Richtlinien-Updates vorformulieren"],roi:"Spart 140 Tsd. $ / Jahr in Support und Suche und erschlieÃt 900 Tsd. $ / Jahr Produktivität.",faq:[{q:"Wo wird Wissen gespeichert?",a:"In Ihrem tenant-isolierten Infernex Vector Store, verschlüsselt und nie für geteiltes Training genutzt."},{q:"Wie oft wird aktualisiert?",a:"Kontinuierliche inkrementelle Synchronisierung, vollständige Neuindexierung auf Wunsch."},{q:"Respektiert er Quellrechte?",a:"Ja â ACLs aus Drive, SharePoint, Notion und Confluence bleiben erhalten."}]},"ai-revenue-intelligence":{name:"KI Revenue Intelligence Agent",tagline:"Ganzheitliche Umsatzintelligenz â Pipeline, Forecast, Risiko.",overview:"AI Revenue Intelligence analysiert CRM-, Billing- und Produktnutzungsdaten, prognostiziert Umsatz, erkennt gefährdete Deals und empfiehlt die nächste beste Vertriebsaktion.",businessValue:["Pipeline in Echtzeit analysiert","Wöchentlicher KI-Umsatzforecast","Frühwarnung für Deal-Risiko und Churn","Kundensegmentierung","Next-best-action pro Account"],workflows:["Pipeline Health Scoring","Wöchentlicher Umsatzforecast","Lost-Deal Post-mortem","Kohortenanalyse","Next-best-action Empfehlung"],roi:"Spart 180 Tsd. $ / Jahr in RevOps und erschlieÃt 2,4 Mio. $ / Jahr zurückgewonnene Pipeline.",faq:[{q:"Welche CRMs?",a:"Salesforce, HubSpot, Pipedrive und Microsoft Dynamics 365 über den Connector Marketplace."},{q:"Wie genau ist der Forecast?",a:"Typisch 92-97% MAPE nach einem Quartal, auf Ihre Historie abgestimmt."},{q:"Kann er CRM-Aktionen ausführen?",a:"Ja â mit menschlicher Freigabe und Audit-Log."}]},"ai-process-automation-auditor":{name:"KI-Prozessautomatisierungs-Auditor",tagline:"Findet jede versteckte Automatisierungschance im Unternehmen.",overview:"AI Process Automation Auditor scannt repetitive Aktivitäten in Workspace, Chat, ERP und CRM und priorisiert Automatisierungen nach geschätztem ROI.",businessValue:["Kontinuierliche Automatisierungserkennung","Heatmap repetitiver Aufgaben","Backlog nach Einsparung sortiert","ROI-Schätzung pro Workflow","Effizienzbenchmark unternehmensweit"],workflows:["Activity Mining und Clustering","Workflow-Diagnose","Backlog und ROI-Ranking","Optimierungsreport für Executives","Ãbergabe an KI-Mitarbeiter"],roi:"Spart 160 Tsd. $ / Jahr in Operations-Analyse und eliminiert 2 Mio. $+ / Jahr repetitive Arbeit.",faq:[{q:"Muss er Dokumente lesen?",a:"StandardmäÃig nur Workflow-Metadaten; Vollzugriff ist optional pro Quelle."},{q:"ERP/CRM-Anbindung?",a:"Ja â SAP, Oracle, NetSuite, Salesforce, HubSpot, Dynamics 365 und Datenbanken."},{q:"Wann gibt es Ergebnisse?",a:"Erstes priorisiertes Backlog innerhalb von 7 Tagen nach Quellenverbindung."}]}
811,"ai-employee-manager":{name:"KI-Mitarbeiter-Manager",tagline:"Verwaltet das gesamte AI Workforce Ãkosystem â Performance, Kosten und Rechte.",overview:"AI Employee Manager verfolgt Leistung, Tokenverbrauch und Kosten jedes KI-Mitarbeiters, empfiehlt Optimierungen, erkennt Fehler, verwaltet Rechte und erstellt CFO/CTO-Reports.",businessValue:["Live-Monitoring jedes KI-Mitarbeiters","Token- und Kosten-Tracking bis Aufgabe","Automatische Modellrouting-Empfehlungen","Fehlererkennung und Auto-Retry","Compliance und Berechtigungen"],workflows:["Performance-Dashboard","Token- und Kostenanalyse","Modelloptimierung","Triage fehlgeschlagener Aufgaben","Berechtigungs- und Audit-Report"],roi:"Spart 120 Tsd. $ / Jahr in AI Ops und 1,5 Mio. $ / Jahr Inferenzkosten.",faq:[{q:"Welche KI-Anbieter werden verfolgt?",a:"Infernex API, OpenAI, Anthropic, Google Gemini, Mistral, Azure AI und eigene Endpoints."},{q:"Kann er Ausgaben begrenzen?",a:"Ja â pro Mitarbeiter, Team und Modell mit Hard Caps und Alerts."},{q:"Ist er erforderlich?",a:"Empfohlen für 3+ KI-Mitarbeiter in Produktion."}]}}},pt:{department:{workforce_command:"Centro de comando IA"},employeeItems:{"ai-workforce-commander":{name:"Comandante AI Workforce (COO IA)",tagline:"Diretor de operações virtual que orquestra toda a sua força de trabalho de IA.",overview:"AI Workforce Commander é a camada central de inteligência operacional da Infernex. Ele monitora cada funcionário de IA, coordena workflows multiagente, detecta gargalos, otimiza processos e gera relatórios executivos em tempo real.",businessValue:["Um COO IA supervisionando toda a força de trabalho","Detecta gargalos entre departamentos","Recomenda automações semanalmente","Relatórios KPI operacionais de nÃvel conselho","Alertas 24/7 sobre anomalias operacionais"],workflows:["Varredura de workflows da empresa","Plano de orquestração multiagente","Relatório de gargalos e agentes ociosos","Briefing executivo de KPI","Backlog de oportunidades de automação"],roi:"Economiza US$ 260 mil/ano em liderança operacional e desbloqueia US$ 3,5 mi/ano em produtividade.",faq:[{q:"Substitui um COO humano?",a:"Não â aumenta a liderança operacional com visibilidade, orquestração e relatórios 24/7."},{q:"Quais funcionários de IA coordena?",a:"Todos os funcionários de IA da Infernex e agentes terceiros conectados via MCP ou API Infernex."},{q:"Os dados operacionais são isolados?",a:"Sim â isolamento por tenant, criptografia AES-256 e trilha de auditoria completa."}]},"ai-knowledge-manager":{name:"Gerente de conhecimento IA",tagline:"Cérebro IA empresarial â cada documento interno respondÃvel instantaneamente.",overview:"AI Knowledge Manager ingere documentos do Drive, SharePoint, Notion e Confluence para criar uma base de conhecimento interna viva com respostas no idioma de cada funcionário.",businessValue:["Respostas instantâneas sobre polÃticas e processos","Documentação obsoleta detectada automaticamente","Assistente multilÃngue para cada equipe","Onboarding 40-60% mais rápido","Sem perda de conhecimento quando funcionários saem"],workflows:["Ingestão e indexação do corpus","Assistente interno de Q&A","Auditoria de atualização documental","Detecção de lacunas de conhecimento","Atualizações de polÃticas pré-redigidas"],roi:"Economiza US$ 140 mil/ano em suporte e busca e desbloqueia US$ 900 mil/ano em produtividade recuperada.",faq:[{q:"Onde o conhecimento fica armazenado?",a:"No vector store Infernex isolado por tenant, criptografado e nunca usado para treinar modelos compartilhados."},{q:"Com que frequência atualiza?",a:"Sincronização incremental contÃnua com reindexação completa sob demanda."},{q:"Respeita permissões da fonte?",a:"Sim â ACLs de Drive, SharePoint, Notion e Confluence são preservadas."}]},"ai-revenue-intelligence":{name:"Agente IA de Revenue Intelligence",tagline:"Inteligência de receita completa â pipeline, previsão e risco.",overview:"AI Revenue Intelligence analisa CRM, faturamento e uso do produto para prever receita, sinalizar negócios em risco e recomendar a próxima melhor ação comercial.",businessValue:["Pipeline analisado em tempo real","Previsão semanal de receita por IA","Alerta precoce de risco e churn","Segmentação de comportamento do cliente","Próxima melhor ação por conta"],workflows:["Scoring de saúde do pipeline","Previsão semanal de receita","Post-mortem de negócios perdidos","Análise de coortes","Recomendador de próxima ação"],roi:"Economiza US$ 180 mil/ano em RevOps e desbloqueia US$ 2,4 mi/ano em pipeline recuperado.",faq:[{q:"Quais CRMs são suportados?",a:"Salesforce, HubSpot, Pipedrive e Microsoft Dynamics 365 via marketplace de conectores."},{q:"Qual a precisão da previsão?",a:"Normalmente 92-97% MAPE após um trimestre, ajustada ao seu histórico."},{q:"Pode enviar ações ao CRM?",a:"Sim, com aprovação humana e auditoria completa."}]},"ai-process-automation-auditor":{name:"Auditor IA de automação de processos",tagline:"Encontra cada oportunidade oculta de automação na empresa.",overview:"AI Process Automation Auditor analisa atividades repetitivas em workspace, chat, ERP e CRM para priorizar automações por ROI estimado.",businessValue:["Descoberta contÃnua de automações","Mapa de calor de tarefas repetitivas","Backlog priorizado por economia","ROI estimado por workflow","Benchmark de eficiência empresarial"],workflows:["Mineração de atividade e clustering","Diagnóstico de workflows ineficientes","Backlog e ranking de ROI","Relatório para executivos","Repasse para funcionários de IA"],roi:"Economiza US$ 160 mil/ano em análise operacional e elimina US$ 2 mi+/ano em trabalho repetitivo.",faq:[{q:"Precisa ler documentos?",a:"Lê metadados por padrão; acesso total ao conteúdo é opcional por fonte."},{q:"Conecta a ERP ou CRM?",a:"Sim â SAP, Oracle, NetSuite, Salesforce, HubSpot, Dynamics 365 e bancos de dados."},{q:"Quando vemos resultados?",a:"Primeiro backlog priorizado em até 7 dias após conectar fontes."}]}
811,"ai-employee-manager":{name:"Gerente de funcionários IA",tagline:"Gerencia todo o ecossistema AI Workforce â desempenho, custo e permissões.",overview:"AI Employee Manager acompanha desempenho, tokens e custo de cada funcionário IA, recomenda otimizações, detecta falhas, gerencia permissões e gera relatórios para CFO e CTO.",businessValue:["Monitoramento ao vivo de cada funcionário IA","Tokens e custos até o nÃvel da tarefa","Recomendações automáticas de roteamento","Detecção de falhas e auto-retry","Conformidade e permissões AI Workforce"],workflows:["Dashboard de desempenho","Análise de tokens e custos","Otimização de modelo","Triagem de tarefas falhas","Relatório de permissões e auditoria"],roi:"Economiza US$ 120 mil/ano em AI Ops e US$ 1,5 mi/ano em custos de inferência.",faq:[{q:"Quais provedores são rastreados?",a:"API Infernex, OpenAI, Anthropic, Google Gemini, Mistral, Azure AI e endpoints customizados."},{q:"Pode impor limites de gasto?",a:"Sim â por funcionário, equipe e modelo, com hard caps e alertas."},{q:"à obrigatório?",a:"Recomendado para 3+ funcionários IA em produção."}]}}}};Xt.zh={department:{workforce_command:"AI ææ¥ä¸å¿"},employeeItems:{"ai-workforce-commander":{name:"AI Workforce ææ¥å®ï¼AI COOï¼",tagline:"èæé¦å¸è¿è¥å®ï¼ç¼æä½ ç宿´ AI åå·¥å¢éã",overview:"AI Workforce Commander æ¯ Infernex çæ ¸å¿è¿è¥æºè½å±ãå®çæ§æ¯ä¸ä¸ªå·²é¨ç½²ç AI åå·¥ï¼åè°å¤ä»£ç工使µï¼åç°ç¶é¢ï¼ä¼åä¸å¡æµç¨ï¼å¹¶ä¸ºç®¡çå±å®æ¶çææ§è¡æ¥åã",businessValue:["ä¸ä¸ª AI COO çç£æ´ä¸ª AI åå·¥å¢é","è·¨é¨é¨åç°å·¥ä½æµç¶é¢","æ¯å¨æ¨èèªå¨åæºä¼","è£äºä¼çº§è¿è¥ KPI æ¥å","7Ã24 å°æ¶è¿è¥å¼å¸¸åè¦"],workflows:["å
¨å
¬å¸å·¥ä½æµæ«æ","å¤ä»£çç¼æè®¡å","ç¶é¢ä¸é²ç½®ä»£çæ¥å","管çå± KPI ç®æ¥","èªå¨åæºä¼ backlog"],roi:"æ¯å¹´èç 26 ä¸ç¾å
è¿è¥é¢å¯¼ææ¬ï¼å¹¶éæ¾æ¯å¹´ 350 ä¸ç¾å
çäº§åæ¶çã",faq:[{q:"å®ä¼æ¿ä»£ç人 COO åï¼",a:"ä¸ä¼ââå®éè¿ 7Ã24 å°æ¶å¯è§åãç¼æåæ¥åå¢å¼ºè¿è¥é¢å¯¼è½åã"},{q:"å®è½åè°åªäº AI åå·¥ï¼",a:"ææ Infernex AI åå·¥ï¼ä»¥åéè¿ MCP æ Infernex API è¿æ¥çç¬¬ä¸æ¹ä»£çã"},{q:"è¿è¥æ°æ®æ¯å¦é离ï¼",a:"æ¯ââç§æ·é离ãAES-256 å å¯å宿´å®¡è®¡è½¨è¿¹ã"}]},"ai-knowledge-manager":{name:"AI ç¥è¯ç®¡çå¨",tagline:"ä¼ä¸ AI 大èââæ¯ä»½å
é¨ææ¡£é½å¯å³æ¶é®çã",overview:"AI Knowledge Manager ä» DriveãSharePointãNotion å Confluence æåå
¬å¸ææ¡£ï¼æå»ºå®æ¶å
é¨ç¥è¯åºãåå·¥å¯ç¨èªå·±çè¯è¨æé®ï¼å¹¶è·å¾åºäºææ°å®æ¹ææ¡£ççæ¡ï¼åæ¶å
ç½®è¿æå
å®¹æ£æµã",businessValue:["峿¶åçå
鍿¿çææµç¨é®é¢","èªå¨æ£æµè¿æææ¡£","é¢åæ¯ä¸ªå¢éçå¤è¯è¨ç¥è¯å©æ","å
¥èæ¶é´åå° 40â60%","åå·¥ç¦»èæ¶é¶ç¥è¯æµå¤±"],workflows:["è¯ææåä¸ç´¢å¼","å
é¨é®ç婿","ææ¡£æ°é²åº¦å®¡è®¡","ç¥è¯ç¼ºå£æ£æµ","èªå¨èµ·èæ¿çæ´æ°"],roi:"æ¯å¹´èç 14 ä¸ç¾å
æ¯æä¸æç´¢ææ¬ï¼å¹¶éæ¾æ¯å¹´ 90 ä¸ç¾å
å¯åæ¶ç产åã",faq:[{q:"ç¥è¯åå¨å¨åªéï¼",a:"åå¨å¨ä½ çç§æ·é离 Infernex åéåºä¸ï¼éæå å¯ï¼ç»ä¸ç¨äºè®ç»å
±äº«æ¨¡åã"},{q:"å
容å¤ä¹
å·æ°ä¸æ¬¡ï¼",a:"æç»å¢é忥ï¼å¹¶å¯æé宿´é建索å¼ã"},{q:"æ¯å¦å°éæ¥æºæéï¼",a:"æ¯ââæ¥è¯¢æ¶ä¿ç DriveãSharePointãNotion å Confluence ç ACLã"}]},"ai-revenue-intelligence":{name:"AI æ¶å
¥æºè½ä»£ç",tagline:"å
¨æ æ¶å
¥æºè½ââ管éã颿µãé£é©ã",overview:"AI Revenue Intelligence æç»åæ CRMãè´¦åå产åä½¿ç¨æ°æ®ï¼é¢æµæ¶å
¥ï¼æ è®°é«é£é©äº¤æï¼å¹¶ä¸ºæ¯ä¸ªè´¦æ·æ¨èä¸ä¸æ¥æä½³éå®è¡å¨ã",businessValue:["宿¶åæéå®ç®¡é","æ¯å¨çæ AI æ¶å
¥é¢æµ","交æé£é©ä¸æµå¤±é¢è¦","客æ·è¡ä¸ºå群","æ¯ä¸ªè´¦æ·æ¨èä¸ä¸æ¥æä½³è¡å¨"],workflows:["管éå¥åº·è¯å","æ¯å¨æ¶å
¥é¢æµ","丢åå¤ç","客æ·è¡ä¸º cohort åæ","ä¸ä¸æ¥æä½³è¡å¨æ¨è"],roi:"æ¯å¹´èç 18 ä¸ç¾å
RevOps ææ¬ï¼å¹¶éæ¾æ¯å¹´ 240 ä¸ç¾å
å¯è¿½å管éä»·å¼ã",faq:[{q:"æ¯æåªäº CRMï¼",a:"SalesforceãHubSpotãPipedrive å Microsoft Dynamics 365ï¼éè¿è¿æ¥å¨å¸åºæ¥å
¥ã"},{q:"颿µæå¤åç¡®ï¼",a:"è¿æ¥åä¸ä¸ªå£åº¦å
éå¸¸è¾¾å° 92â97% MAPEï¼å¹¶åºäºä½ çå岿°æ®è°ä¼ã"},{q:"è½å CRM æ¨éå¨ä½åï¼",a:"å¯ä»¥ââ带人工审æ¹å·¥ä½æµå宿´å®¡è®¡æ¥å¿ã"}]},"ai-process-automation-auditor":{name:"AI æµç¨èªå¨å审计å",tagline:"åç°å
¬å¸å
鍿¯ä¸ä¸ªéèçèªå¨åæºä¼ã",overview:"AI Process Automation Auditor æ«æ Google Worksp
811aceãMicrosoft 365ãè天ãERP å CRM ä¸çé夿´»å¨ï¼æé¢è®¡èçææ¬ä¼å
æåºæé« ROI çèªå¨åæºä¼ã",businessValue:["æç»åç°èªå¨åæºä¼","æå¢éå±ç¤ºéå¤ä»»å¡çåå¾","æèçé颿åºçèªå¨å backlog","æ¯ä¸ªå·¥ä½æµèªå¨ä¼°ç® ROI","ä¼ä¸çº§æçåºå"],workflows:["æ´»å¨ææä¸ä»»å¡èç±»","使工使µè¯æ","èªå¨å backlog ä¸ ROI æå","é¢å管çå±çä¼åæ¥å","ç§»äº¤ç» AI åå·¥æ§è¡"],roi:"æ¯å¹´èç 16 ä¸ç¾å
è¿è¥åæææ¬ï¼å¹¶æ¶é¤æ¯å¹´ 200 ä¸ç¾å
以ä¸éå¤å·¥ä½ã",faq:[{q:"å®éè¦è¯»åæä»¬çææ¡£åï¼",a:"é»è®¤è¯»å工使µå
æ°æ®ï¼å®æ´å
å®¹è®¿é®ææ¥æºéæ©å¼å¯ã"},{q:"è½æ¥å
¥ ERP æ CRM åï¼",a:"å¯ä»¥ââSAPãOracleãNetSuiteãSalesforceãHubSpotãDynamics 365 以忰æ®åºã"},{q:"å¤ä¹
è½çå°ç»æï¼",a:"è¿æ¥æ°æ®æºå 7 天å
交ä»ç¬¬ä¸ä»½ä¼å
级èªå¨å backlogã"}]},"ai-employee-manager":{name:"AI å工管çå¨",tagline:"ç®¡çæ´ä¸ª AI Workforce çæââæ§è½ãææ¬åæéã",overview:"AI Employee Manager è·è¸ªæ¯ä¸ª AI åå·¥çæ§è½ãtoken æ¶èåææ¬ã宿¨è模åä¼åï¼æ 记失败任å¡ï¼ç®¡çæéï¼å¹¶ä¸º CFO å CTO çæå®æ´ AI Workforce æ¥åã",businessValue:["宿¶çæ§æ¯ä¸ª AI åå·¥","精确å°ä»»å¡ç token 䏿æ¬è¿½è¸ª","èªå¨æ¨¡åè·¯ç±å»ºè®®","å¤±è´¥ä»»å¡æ£æµä¸èªå¨éè¯çç¥","AI Workforce åè§ä¸æé管ç"],workflows:["AI åå·¥æ§è½ä»ªè¡¨ç","Token æ¶è䏿æ¬åæ","模åä¼å建议","失败任å¡éååè¯","æéä¸å®¡è®¡æ¥å"],roi:"æ¯å¹´èç 12 ä¸ç¾å
AI Ops ææ¬ï¼å¹¶éæ¾æ¯å¹´ 150 ä¸ç¾å
æ¨çææ¬èçã",faq:[{q:"追踪åªäº AI æä¾åï¼",a:"Infernex APIãOpenAIãAnthropicãGoogle GeminiãMistralãAzure AI åä»»ä½èªå®ä¹ endpointã"},{q:"è½å¼ºå¶æ§è¡æ¯åºéå¶åï¼",a:"å¯ä»¥ââæåå·¥ãå¢éãæ¨¡å设置 hard cap ä¸åè¦ã"},{q:"å
¶ä» AI åå·¥å¿
须使ç¨å®åï¼",a:"建议任ä½è¿è¡ 3 个以ä¸ç产 AI åå·¥çå¢é使ç¨ã"}]}}};Xt.ja={department:{workforce_command:"AIã³ãã³ãã»ã³ã¿ã¼"},employeeItems:{"ai-workforce-commander":{name:"AI Workforce Commanderï¼AI COOï¼",tagline:"AIã¯ã¼ã¯ãã©ã¼ã¹å
¨ä½ãçµ±æ¬ããä»®æ³æé«å·è¡è²¬ä»»è
ã",overview:"AI Workforce Commander 㯠Infernex ã®ä¸æ ¸ã¨ãªãæ¥åã¤ã³ããªã¸ã§ã³ã¹å±¤ã§ããå±éæ¸ã¿ã®AI徿¥å¡ãç£è¦ãããã«ãã¨ã¼ã¸ã§ã³ãã®ã¯ã¼ã¯ããã¼ã調æ´ããããã«ããã¯ãæ¤åºããæ¥åããã»ã¹ãæé©åããçµå¶å±¤åãã¬ãã¼ãããªã¢ã«ã¿ã¤ã ã§çæãã¾ãã",businessValue:["AIã¯ã¼ã¯ãã©ã¼ã¹å
¨ä½ãç£ç£ããAI COO","é¨é横æã®ããã«ããã¯ãæ¤åº","æ¯é±èªååæ©ä¼ãææ¡","åç· å½¹ä¼ã¬ãã«ã®éç¨KPIã¬ãã¼ã","éç¨ç°å¸¸ã24æé365æ¥ã¢ã©ã¼ã"],workflows:["å
¨ç¤¾ã¯ã¼ã¯ããã¼ã¹ãã£ã³","ãã«ãã¨ã¼ã¸ã§ã³ãçµ±æ¬è¨ç»","ããã«ããã¯ã¨ä¼æ¢ã¨ã¼ã¸ã§ã³ãã®ã¬ãã¼ã","çµå¶KPIããªã¼ãã£ã³ã°","èªååæ©ä¼ããã¯ãã°"],roi:"éç¨ãªã¼ãã¼ã·ããã§å¹´é26ä¸ãã«ã忏ããå¹´é350ä¸ãã«ã®çç£æ§åä¸ãå®ç¾ã",faq:[{q:"人éã®COOãç½®ãæãã¾ããï¼",a:"ãããã24æé365æ¥ã®å¯è¦åãçµ±æ¬ãã¬ãã¼ãã§éç¨ãªã¼ãã¼ã·ãããå¼·åãã¾ãã"},{q:"ã©ã®AI徿¥å¡ã調æ´ã§ãã¾ããï¼",a:"ãã¹ã¦ã®Infernex AI徿¥å¡ã¨ãMCPã¾ãã¯Infernex APIã§æ¥ç¶ãããå¤é¨ã¨ã¼ã¸ã§ã³ãã§ãã"},{q:"éç¨ãã¼ã¿ã¯åé¢ããã¦ãã¾ããï¼",a:"ã¯ããããã³ãåé¢ãAES-256æå·åãå®å
¨ãªç£æ»è¨¼è·¡ãããã¾ãã"}]},"ai-knowledge-manager":{name:"AIãã¬ãã¸ããã¼ã¸ã£ã¼",tagline:"伿¥AIãã¬ã¤ã³ â ãã¹ã¦ã®å
鍿æ¸ã峿åçå¯è½ã«ã",overview:"AI Knowledge Manager 㯠DriveãSharePointãNotionãConfluence ããææ¸ãåãè¾¼ã¿ãææ°ã®å
¬å¼ææ¸ã«åºã¥ãã©ã¤ã社å
ãã¬ãã¸ãã¼ã¹ãæ§ç¯ãã¾ãã",businessValue:["社å
ããªã·ã¼ãããã»ã¹ã¸å³æåç","å¤ãææ¸ãèªåæ¤åº","å
¨ãã¼ã åãå¤è¨èªãã¬ãã¸å©æ","ãªã³ãã¼ãã£ã³ã°æéã40ã60%ç縮","éè·æã®ç¥èåªå¤±ã鲿¢"],workflows:["ã³ã¼ãã¹åãè¾¼ã¿ã¨ç´¢å¼å","社å
Q&Aã¢ã·ã¹ã¿ã³ã","ææ¸é®®åº¦ç£æ»","ç¥èã®ã£ããæ¤åº","ããªã·ã¼æ´æ°æ¡ã®èªå使"],roi:"ãµãã¼ãã¨æ¤ç´¢ã§å¹´é14ä¸ãã«ã忏ããå¹´é90ä¸ãã«ã®çç£æ§ãååã",faq:[{q:"ç¥èã¯ã©ãã«ä¿åãã
811ã¾ããï¼",a:"ããã³ãåé¢ãããInfernexãã¯ãã«ã¹ãã¢ã«ä¿åãããæå·åãããå
±æã¢ãã«ã®å¦ç¿ã«ã¯ä½¿ããã¾ããã"},{q:"æ´æ°é »åº¦ã¯ï¼",a:"ç¶ç¶çãªå·®ååæã¨ãå¿
è¦ã«å¿ããå®å
¨åç´¢å¼åã§ãã"},{q:"å
ã®æ¨©éãå°éãã¾ããï¼",a:"ã¯ããDriveãSharePointãNotionãConfluence ã®ACLãã¯ã¨ãªæã«ä¿æãã¾ãã"}]},"ai-revenue-intelligence":{name:"AI Revenue Intelligence ã¨ã¼ã¸ã§ã³ã",tagline:"売ä¸ã¤ã³ããªã¸ã§ã³ã¹ â ãã¤ãã©ã¤ã³ãäºæ¸¬ããªã¹ã¯ã",overview:"AI Revenue Intelligence ã¯CRMãè«æ±ã製åå©ç¨ãã¼ã¿ãç¶ç¶åæããåçãäºæ¸¬ãããªã¹ã¯æ¡ä»¶ãæ¤åºããåã¢ã«ã¦ã³ãã®æ¬¡åã¢ã¯ã·ã§ã³ãæ¨å¥¨ãã¾ãã",businessValue:["å¶æ¥ãã¤ãã©ã¤ã³ããªã¢ã«ã¿ã¤ã åæ","æ¯é±AIåçäºæ¸¬ãçæ","æ¡ä»¶ãªã¹ã¯ã¨ãã£ã¼ã³ã®æ©æè¦å","顧客è¡åã»ã°ã¡ã³ãå","ã¢ã«ã¦ã³ãå¥ã®æ¬¡åã¢ã¯ã·ã§ã³"],workflows:["ãã¤ãã©ã¤ã³å¥å
¨æ§ã¹ã³ã¢","鱿¬¡åçäºæ¸¬","失注åæ","顧客è¡åã³ãã¼ãåæ","次åã¢ã¯ã·ã§ã³æ¨è¦"],roi:"RevOpsã§å¹´é18ä¸ãã«ã忏ããå¹´é240ä¸ãã«ã®ãã¤ãã©ã¤ã³ãå復ã",faq:[{q:"対å¿CRMã¯ï¼",a:"SalesforceãHubSpotãPipedriveãMicrosoft Dynamics 365 ãã³ãã¯ã¿ãã¼ã±ãããã¬ã¤ã¹çµç±ã§æ¥ç¶ã§ãã¾ãã"},{q:"äºæ¸¬ç²¾åº¦ã¯ï¼",a:"æ¥ç¶å¾1ååæã§é常92ã97% MAPEã貴社履æ´ã«åããã¦èª¿æ´ããã¾ãã"},{q:"CRMã¸ã¢ã¯ã·ã§ã³ãéãã¾ããï¼",a:"ã¯ãã人éã®æ¿èªã¯ã¼ã¯ããã¼ã¨å®å
¨ãªç£æ»ãã°ä»ãã§ãã"}]},"ai-process-automation-auditor":{name:"AIããã»ã¹èªååç£æ»å¡",tagline:"社å
ã®é ããèªååæ©ä¼ããã¹ã¦çºè¦ã",overview:"AI Process Automation Auditor 㯠workspaceããã£ãããERPãCRM ã®å復æ¥åãåæããæ¨å®ROIã®é«ãèªååæ©ä¼ãåªå
é ä½ä»ããã¾ãã",businessValue:["èªååæ©ä¼ãç¶ç¶çã«çºè¦","ãã¼ã å¥ã®å復ã¿ã¹ã¯ãã¼ãããã","忏é¡é ã®èªååããã¯ãã°","ã¯ã¼ã¯ããã¼å¥ROIæ¨å®","å
¨ç¤¾å¹çãã³ããã¼ã¯"],workflows:["æ´»åãã¤ãã³ã°ã¨ã¯ã©ã¹ã¿ãªã³ã°","éå¹çã¯ã¼ã¯ããã¼è¨ºæ","ããã¯ãã°ã¨ROIã©ã³ãã³ã°","çµå¶å±¤åãæé©åã¬ãã¼ã","AI徿¥å¡ã¸ã®å®è¡å¼ãç¶ã"],roi:"æ¥ååæã§å¹´é16ä¸ãã«ã忏ããå¹´é200ä¸ãã«è¶
ã®å復使¥ã忏ã",faq:[{q:"ææ¸ãèªãå¿
è¦ãããã¾ããï¼",a:"æ¢å®ã§ã¯ã¯ã¼ã¯ããã¼ã¡ã¿ãã¼ã¿ã®ã¿ãå
¨æã¢ã¯ã»ã¹ã¯ã½ã¼ã¹ãã¨ã«ä»»æã§ãã"},{q:"ERPãCRMã«æ¥ç¶ã§ãã¾ããï¼",a:"ã¯ããSAPãOracleãNetSuiteãSalesforceãHubSpotãDynamics 365ãå種DBã«å¯¾å¿ãã¾ãã"},{q:"çµæã¯ãã¤åºã¾ããï¼",a:"ãã¼ã¿ã½ã¼ã¹æ¥ç¶å¾7æ¥ä»¥å
ã«æåã®åªå
ããã¯ãã°ãæä¾ãã¾ãã"}]},"ai-employee-manager":{name:"AI徿¥å¡ããã¼ã¸ã£ã¼",tagline:"AI Workforceå
¨ä½ã®æ§è½ãã³ã¹ããæ¨©éã管çã",overview:"AI Employee Manager ã¯åAI徿¥å¡ã®æ§è½ããã¼ã¯ã³æ¶è²»ãã³ã¹ãã追跡ããã¢ãã«æé©åã失æã¿ã¹ã¯æ¤åºã権é管çãCFO/CTOåãã¬ãã¼ããè¡ãã¾ãã",businessValue:["åAI徿¥å¡ã®ã©ã¤ãç£è¦","ã¿ã¹ã¯åä½ã®ãã¼ã¯ã³ã¨ã³ã¹ã追跡","ã¢ãã«ã«ã¼ãã£ã³ã°èªåææ¡","失æã¿ã¹ã¯æ¤åºã¨èªåãªãã©ã¤","AI Workforceã®ã³ã³ãã©ã¤ã¢ã³ã¹ã¨æ¨©é"],workflows:["AI徿¥å¡æ§è½ããã·ã¥ãã¼ã","ãã¼ã¯ã³æ¶è²»ã¨ã³ã¹ãåæ","ã¢ãã«æé©åææ¡","失æã¿ã¹ã¯ãã¥ã¼ã®ããªã¢ã¼ã¸","権éã¨ç£æ»ã¬ãã¼ã"],roi:"AI Opsã§å¹´é12ä¸ãã«ã忏ããå¹´é150ä¸ãã«ã®æ¨è«ã³ã¹ããç¯ç´ã",faq:[{q:"ã©ã®AIãããã¤ãã¼ã追跡ãã¾ããï¼",a:"Infernex APIãOpenAIãAnthropicãGoogle GeminiãMistralãAzure AIãä»»æã®ã«ã¹ã¿ã endpointã"},{q:"æ¯åºä¸éãå¼·å¶ã§ãã¾ããï¼",a:"ã¯ãã徿¥å¡ããã¼ã ãã¢ãã«ãã¨ã«ãã¼ããã£ããã¨ã¢ã©ã¼ããè¨å®ã§ãã¾ãã"},{q:"å¿
é ã§ããï¼",a:"æ¬çªã§3ã¤ä»¥ä¸ã®AI徿¥å¡ãéç¨ããå ´åã«æ¨å¥¨ãã¾ãã"}]}}};Xt.hi={department:{workforce_command:"AI à¤à¤®à¤¾à¤à¤¡ सà¥à¤à¤à¤°"}
811,employeeItems:{"ai-workforce-commander":{name:"AI Workforce Commander (AI COO)",tagline:"à¤à¤ªà¤à¥ पà¥à¤°à¥ AI workforce à¤à¥ orchestrate à¤à¤°à¤¨à¥ वाला virtual Chief Operating Officer.",overview:"AI Workforce Commander Infernex à¤à¥ à¤à¥à¤à¤¦à¥à¤°à¥à¤¯ operational intelligence layer हà¥à¥¤ यह हर deployed AI Employee à¤à¥ monitor à¤à¤°à¤¤à¤¾ हà¥, multi-agent workflows coordinate à¤à¤°à¤¤à¤¾ हà¥, bottlenecks detect à¤à¤°à¤¤à¤¾ हà¥, business processes optimize à¤à¤°à¤¤à¤¾ हॠà¤à¤° C-suite à¤à¥ लिठreal-time executive reports बनाता हà¥à¥¤",businessValue:["पà¥à¤°à¥ AI workforce à¤à¥ दà¥à¤à¤°à¥à¤ à¤à¤°à¤¨à¥ वाला à¤à¤ AI COO","Departments à¤à¥ बà¥à¤ workflow bottlenecks detect à¤à¤°à¤¤à¤¾ हà¥","हर सपà¥à¤¤à¤¾à¤¹ automation opportunities सà¥à¤à¤¾à¤¤à¤¾ हà¥","Board-grade operational KPI reporting","Operational anomalies पर 24/7 alerts"],workflows:["Company-wide workflow scan","Multi-agent orchestration plan","Bottleneck à¤à¤° idle-agent report","Executive KPI briefing","Automation opportunity backlog"],roi:"Operations leadership मà¥à¤ $260k/वरà¥à¤· बà¤à¤¤ à¤à¤° productivity मà¥à¤ $3.5M/वरà¥à¤· unlock à¤à¤°à¤¤à¤¾ हà¥à¥¤",faq:[{q:"à¤à¥à¤¯à¤¾ यह human COO à¤à¥ replace à¤à¤°à¤¤à¤¾ हà¥?",a:"नहà¥à¤ â यह 24/7 visibility, orchestration à¤à¤° reporting सॠoperational leadership à¤à¥ augment à¤à¤°à¤¤à¤¾ हà¥à¥¤"},{q:"यह à¤à¥à¤¨ सॠAI Employees coordinate à¤à¤° सà¤à¤¤à¤¾ हà¥?",a:"सà¤à¥ Infernex AI Employees à¤à¤° MCP या Infernex API सॠà¤à¥à¤¡à¤¼à¥ third-party agents."},{q:"à¤à¥à¤¯à¤¾ operational data isolated हà¥?",a:"हाठâ tenant isolation, AES-256 encryption à¤à¤° full audit trail."}]},"ai-knowledge-manager":{name:"AI Knowledge Manager",tagline:"Enterprise AI brain â हर internal document तà¥à¤°à¤à¤¤ answerable.",overview:"AI Knowledge Manager Drive, SharePoint, Notion à¤à¤° Confluence सॠcompany documents ingest à¤à¤°à¤à¥ live internal knowledge base बनाता हà¥à¥¤ Employees à¤
पनॠà¤à¤¾à¤·à¤¾ मà¥à¤ सवाल पà¥à¤à¤¤à¥ हà¥à¤ à¤à¤° latest official documentation पर à¤à¤§à¤¾à¤°à¤¿à¤¤ answers पातॠहà¥à¤à¥¤",businessValue:["à¤à¤¿à¤¸à¥ à¤à¥ internal policy या process पर instant answers","Outdated documentation auto-detect","हर team à¤à¥ लिठmultilingual knowledge assistant","Onboarding time 40-60% à¤à¤®","Employees leave à¤à¤°à¤¨à¥ पर zero knowledge loss"],workflows:["Corpus ingestion à¤à¤° indexing","Internal Q&A assistant","Documentation freshness audit","Knowledge gap detection","Auto-drafted policy updates"],roi:"Support à¤à¤° search मà¥à¤ $140k/वरà¥à¤· बà¤à¤¤ à¤à¤° reclaimed productivity मà¥à¤ $900k/वरà¥à¤· unlock à¤à¤°à¤¤à¤¾ हà¥à¥¤",faq:[{q:"Knowledge à¤à¤¹à¤¾à¤ stored हà¥?",a:"à¤à¤ªà¤à¥ tenant-isolated Infernex vector store मà¥à¤ â encrypted at rest, shared models train à¤à¤°à¤¨à¥ à¤à¥ लिठà¤à¤à¥ à¤à¤ªà¤¯à¥à¤ नहà¥à¤à¥¤"},{q:"Content à¤à¤¿à¤¤à¤¨à¥ बार refresh हà¥à¤¤à¤¾ हà¥?",a:"Continuous incremental sync à¤à¤° on-demand full re-index."},{q:"à¤à¥à¤¯à¤¾ source permissions respect हà¥à¤¤à¥ हà¥à¤?",a:"हाठâ Drive, SharePoint, Notion à¤à¤° Confluence à¤à¥ ACLs query time पर preserved रहतॠहà¥à¤à¥¤"}]},"ai-revenue-intelligence":{name:"AI Revenue Intelligence Agent",tagline:"Full-stack revenue intelligence â pipeline, forecast, risk.",overview:"AI Revenue Intelligence CRM, billing à¤à¤° product-usage data à¤à¥ continuously analyse à¤à¤°à¤à¥ revenue forecast à¤à¤°à¤¤à¤¾ हà¥, at-risk deals flag à¤à¤°à¤¤à¤¾ हॠà¤à¤° हर account à¤à¥ लिठnext best sales action recommend à¤à¤°à¤¤à¤¾ हà¥à¥¤",businessValue:["Sales pipeline real time मà¥à¤ analysed","हर सपà¥à¤¤à¤¾à¤¹ AI-generated revenue forecast","Deal-risk à¤à¤° churn early warning","Customer behaviour segmentation","हर account à¤à¥ लिठnext best action"],workflows:["Pipeline health scoring","Weekly revenue forecast","Lost-deal post-mortem","Customer behaviour cohort analysis","Next-best-action recommender"],roi:"RevOps मà¥à¤ $180k/वरà¥à¤· बà¤à¤¤ à¤à¤° recovered pipeline मà¥à¤ $2.4M/वरà¥à¤· unlock à¤à¤°à¤¤à¤¾ हà¥à¥¤",faq:[{q:"à¤à¥à¤¨ सॠCRMs supported हà¥à¤?",a:"Salesforce, HubSpot, Pipedrive à¤à¤° Microsoft Dynamics 365 â Connector Marketplace à¤à¥ à¤à¤¼à¤°à¤¿à¤¯à¥à¥¤"},{q:"Forecast à¤à¤¿à¤¤à¤¨à¤¾ accurate हà¥?",a:"Connection à¤à¥ à¤à¤ quarter बाद सामानà¥à¤¯à¤¤à¤ 92-97% MAPE, à¤à¤ªà¤à¥ history पर tuned."},{q:"à¤à¥à¤¯à¤¾ यह CRM मà¥à¤ actions push à¤à¤° सà¤à¤¤à¤¾ हà¥?",a:"हाठâ human approval workflows à¤à¤° full audit log à¤à¥ साथ."}]},"ai-process-automation-auditor":{name:"AI Process Automation Auditor",tagline:"à¤à¤ªà¤à¥ company à¤à¥ हर hidden automation opportunity à¤à¥ à¤à¥à¤à¤¤à¤¾ हà¥à¥¤",overview:"AI Process Automation Auditor Google Worksp
811ace, Microsoft 365, chat, ERP à¤à¤° CRM मà¥à¤ repetitive activity scan à¤à¤°à¤à¥ estimated saved cost à¤à¥ à¤à¤§à¤¾à¤° पर high-ROI automation opportunities prioritize à¤à¤°à¤¤à¤¾ हà¥à¥¤",businessValue:["Continuous automation opportunity discovery","Team-wise repetitive-task heatmap","Cost savings à¤à¥ साथ ranked automation backlog","हर workflow पर zero-effort ROI estimation","Enterprise-wide efficiency benchmark"],workflows:["Activity mining à¤à¤° task clustering","Inefficient workflow diagnosis","Automation backlog à¤à¤° ROI ranking","C-suite optimization report","Execution à¤à¥ लिठAI Employees à¤à¥ handover"],roi:"Ops analysis मà¥à¤ $160k/वरà¥à¤· बà¤à¤¤ à¤à¤° repetitive work elimination मà¥à¤ $2M+/वरà¥à¤· unlock à¤à¤°à¤¤à¤¾ हà¥à¥¤",faq:[{q:"à¤à¥à¤¯à¤¾ à¤à¤¸à¥ documents पढ़नॠपड़तॠहà¥à¤?",a:"Default मà¥à¤ workflow metadata पढ़ता हà¥à¥¤ Full content access हर source पर opt-in हà¥à¥¤"},{q:"à¤à¥à¤¯à¤¾ यह ERP या CRM सॠà¤à¥à¤¡à¤¼à¤¤à¤¾ हà¥?",a:"हाठâ SAP, Oracle, NetSuite, Salesforce, HubSpot, Dynamics 365 à¤à¤° databases."},{q:"Results à¤à¤¬ दिà¤à¤¤à¥ हà¥à¤?",a:"Sources connect à¤à¤°à¤¨à¥ à¤à¥ 7 दिनà¥à¤ मà¥à¤ पहला ranked automation backlog."}]},"ai-employee-manager":{name:"AI Employee Manager",tagline:"पà¥à¤°à¥ AI workforce ecosystem à¤à¥ manage à¤à¤°à¥à¤ â performance, cost à¤à¤° permissions.",overview:"AI Employee Manager हर AI Employee à¤à¥ performance, token consumption à¤à¤° cost track à¤à¤°à¤¤à¤¾ हà¥à¥¤ यह model optimizations recommend à¤à¤°à¤¤à¤¾ हà¥, failed tasks flag à¤à¤°à¤¤à¤¾ हà¥, permissions manage à¤à¤°à¤¤à¤¾ हॠà¤à¤° CFO/CTO à¤à¥ लिठfull AI workforce report बनाता हà¥à¥¤",businessValue:["हर AI Employee à¤à¥ live performance monitoring","Task level तठtoken à¤à¤° cost tracking","Automatic model routing recommendations","Failed-task detection à¤à¤° auto-retry policies","AI workforce compliance à¤à¤° permissions"],workflows:["AI Employee performance dashboard","Token consumption à¤à¤° cost analysis","Model optimization recommendation","Failed-task queue triage","Permissions à¤à¤° audit report"],roi:"AI Ops मà¥à¤ $120k/वरà¥à¤· बà¤à¤¤ à¤à¤° inference cost savings मà¥à¤ $1.5M/वरà¥à¤· unlock à¤à¤°à¤¤à¤¾ हà¥à¥¤",faq:[{q:"à¤à¥à¤¨ सॠAI providers tracked हà¥à¤?",a:"Infernex API, OpenAI, Anthropic, Google Gemini, Mistral, Azure AI à¤à¤° custom endpoints."},{q:"à¤à¥à¤¯à¤¾ यह spend limits enforce à¤à¤° सà¤à¤¤à¤¾ हà¥?",a:"हाठâ employee, team à¤à¤° model level पर hard caps à¤à¤° alerting."},{q:"à¤à¥à¤¯à¤¾ यह required हà¥?",a:"Production मà¥à¤ 3+ AI Employees à¤à¤²à¤¾à¤¨à¥ पर recommended."}]}}};Xt.ar={department:{workforce_command:"Ù
Ø±ÙØ² ÙÙØ§Ø¯Ø© Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ"},employeeItems:{"ai-workforce-commander":{name:"ÙØ§Ø¦Ø¯ ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·Ùاع٠(COO AI)",tagline:"Ù
Ø¯ÙØ± عÙ
ÙÙØ§Øª Ø§ÙØªØ±Ø§Ø¶Ù ÙÙØ³ÙÙ ÙØ§Ù
Ù ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ.",overview:"AI Workforce Commander ÙÙ Ø·Ø¨ÙØ© Ø§ÙØ°Ùاء Ø§ÙØªØ´ØºÙÙ٠اÙÙ
Ø±ÙØ²ÙØ© ÙÙ Infernex. ÙØ±Ø§Ùب ÙÙ Ù
ÙØ¸Ù AI Ù
ÙØ´ÙØ±Ø ÙÙØ³ÙÙ Ø³ÙØ± Ø§ÙØ¹Ù
Ù Ù
تعدد اÙÙÙÙØ§Ø¡Ø ÙÙØªØ´Ù Ø§ÙØ§Ø®ØªÙØ§ÙØ§ØªØ ÙØØ³ÙÙ Ø§ÙØ¹Ù
ÙÙØ§ØªØ ÙÙÙØ´Ø¦ ØªÙØ§Ø±Ùر تÙÙÙØ°ÙØ© ÙÙØ¥Ø¯Ø§Ø±Ø© Ø§ÙØ¹ÙÙØ§ Ù٠اÙÙÙØª Ø§ÙØÙÙÙÙ.",businessValue:["COO AI ÙØ§ØØ¯ ÙØ´Ø±Ù عÙÙ ÙØ§Ù
Ù ÙÙØ© Ø§ÙØ¹Ù
Ù","ÙÙØªØ´Ù Ø§ÙØ§Ø®ØªÙØ§ÙØ§Øª بÙÙ Ø§ÙØ£ÙساÙ
","ÙÙØªØ±Ø ÙØ±Øµ أتÙ
تة Ø£Ø³Ø¨ÙØ¹ÙØ©","ØªÙØ§Ø±Ùر KPI تشغÙÙÙØ© بÙ
ستÙÙ Ù
Ø¬ÙØ³ Ø§ÙØ¥Ø¯Ø§Ø±Ø©","ØªÙØ¨ÙÙØ§Øª 24/7 عÙÙ Ø§ÙØ´Ø°Ùذ Ø§ÙØªØ´ØºÙÙÙ"],workflows:["Ù
Ø³Ø Ø³ÙØ± Ø§ÙØ¹Ù
٠عÙÙ Ù
ستÙÙ Ø§ÙØ´Ø±ÙØ©","خطة ØªÙØ³ÙÙ Ù
تعددة اÙÙÙÙØ§Ø¡","ØªÙØ±Ùر Ø§ÙØ§Ø®ØªÙØ§ÙØ§Øª ÙØ§ÙÙÙÙØ§Ø¡ Ø§ÙØ®Ø§Ù
ÙÙÙ","Ù
ÙØ¬Ø² KPI تÙÙÙØ°Ù","ÙØ§Ø¦Ù
Ø© ÙØ±Øµ Ø§ÙØ£ØªÙ
تة"],roi:"ÙÙÙØ± 260 Ø£Ù٠دÙÙØ§Ø± سÙÙÙÙØ§ Ù٠اÙÙÙØ§Ø¯Ø© Ø§ÙØªØ´ØºÙÙÙØ© ÙÙÙØªØ 3.5 Ù
ÙÙÙ٠دÙÙØ§Ø± سÙÙÙÙØ§ Ù
Ù Ù
ÙØ§Ø³Ø¨ Ø§ÙØ¥ÙØªØ§Ø¬ÙØ©.",faq:[{q:"ÙÙ ÙØ³ØªØ¨Ø¯Ù COO Ø§ÙØ¨Ø´Ø±ÙØ",a:"ÙØ§ â ÙØ¹Ø²Ø² اÙÙÙØ§Ø¯Ø© Ø§ÙØªØ´ØºÙÙÙØ© Ø¨Ø±Ø¤ÙØ© ÙØªÙسÙÙ ÙØªÙØ§Ø±ÙØ± 24/7."},{q:"Ø£Ù Ù
ÙØ¸ÙÙ AI ÙÙ
ÙÙÙ ØªÙØ³ÙÙÙÙ
Ø",a:"ÙÙ Ù
ÙØ¸ÙÙ Infernex AIØ Ø¥Ø¶Ø§ÙØ© Ø¥Ù٠اÙÙÙÙØ§Ø¡ Ø§ÙØ®Ø§Ø±Ø¬ÙÙ٠اÙÙ
تصÙÙ٠عبر MCP Ø£Ù API Infernex."},{q:"ÙÙ Ø§ÙØ¨ÙØ§ÙØ§Øª Ø§ÙØªØ´ØºÙÙÙØ© Ù
عزÙÙØ©Ø",a:"ÙØ¹Ù
â Ø¹Ø²Ù ØØ³Ø¨ اÙÙ
Ø³ØªØ£Ø¬Ø±Ø ØªØ´ÙÙØ± AES-256Ø ÙÙ
سار تدÙÙÙ ÙØ§Ù
Ù."}]}
811,"ai-knowledge-manager":{name:"Ù
Ø¯ÙØ± اÙÙ
Ø¹Ø±ÙØ© Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",tagline:"عÙÙ AI Ù
ؤسس٠â ÙÙ Ù
Ø³ØªÙØ¯ داخÙÙ ÙØ§Ø¨Ù ÙÙØ¥Ø¬Ø§Ø¨Ø© ÙÙØ±Ùا.",overview:"AI Knowledge Manager ÙØ³ØªÙعب Ù
Ø³ØªÙØ¯Ø§Øª Drive ÙSharePoint ÙNotion ÙConfluence ÙØ¨Ùاء ÙØ§Ø¹Ø¯Ø© Ù
Ø¹Ø±ÙØ© داخÙÙØ© ØÙØ©. ÙØ³Ø£Ù اÙÙ
ÙØ¸ÙÙÙ Ø¨ÙØºØªÙÙ
ÙÙØØµÙÙ٠عÙ٠إجابات Ù
Ø³ØªÙØ¯Ø© Ø¥ÙÙ Ø£ØØ¯Ø« اÙÙØ«Ø§Ø¦Ù Ø§ÙØ±Ø³Ù
ÙØ© Ù
ع ÙØ´Ù اÙÙ
ØØªÙ٠اÙÙØ¯ÙÙ
.",businessValue:["إجابات ÙÙØ±ÙØ© Ø¹Ù Ø§ÙØ³Ùاسات ÙØ§ÙعÙ
ÙÙØ§Øª","ÙØ´Ù تÙÙØ§Ø¦Ù ÙÙÙØ«Ø§Ø¦Ù اÙÙØ¯ÙÙ
Ø©","Ù
ساعد Ù
Ø¹Ø±ÙØ© Ù
تعدد اÙÙØºØ§Øª ÙÙÙ ÙØ±ÙÙ","تÙÙÙÙ ÙÙØª onboarding Ø¨ÙØ³Ø¨Ø© 40-60%","ÙØ§ ÙÙØ¯Ø§Ù Ù
Ø¹Ø±ÙØ© Ø¹ÙØ¯ Ù
غادرة اÙÙ
ÙØ¸ÙÙÙ"],workflows:["Ø§Ø³ØªÙØ¹Ø§Ø¨ ÙÙÙØ±Ø³Ø© اÙÙ
ØØªÙÙ","Ù
ساعد Ø£Ø³Ø¦ÙØ© داخÙÙ","تدÙÙÙ ØØ¯Ø§Ø«Ø© اÙÙØ«Ø§Ø¦Ù","ÙØ´Ù ÙØ¬Ùات اÙÙ
Ø¹Ø±ÙØ©","ØµÙØ§ØºØ© ØªØØ¯Ùثات Ø§ÙØ³Ùاسات تÙÙØ§Ø¦ÙÙØ§"],roi:"ÙÙÙØ± 140 Ø£Ù٠دÙÙØ§Ø± سÙÙÙÙØ§ ÙÙ Ø§ÙØ¯Ø¹Ù
ÙØ§ÙØ¨ØØ« ÙÙÙØªØ 900 Ø£Ù٠دÙÙØ§Ø± سÙÙÙÙØ§ Ù
Ù Ø§ÙØ¥ÙØªØ§Ø¬ÙØ© اÙÙ
ستعادة.",faq:[{q:"Ø£ÙÙ ØªÙØ®Ø²Ù اÙÙ
Ø¹Ø±ÙØ©Ø",a:"ÙÙ vector store Ù
عزÙÙ ÙÙ
ؤسست٠داخ٠InfernexØ Ù
Ø´ÙØ± ÙÙØ§ ÙÙØ³ØªØ®Ø¯Ù
ÙØªØ¯Ø±Ùب ÙÙ
اذج Ù
Ø´ØªØ±ÙØ©."},{q:"ÙÙ
Ù
رة ÙØªÙ
Ø§ÙØªØØ¯ÙØ«Ø",a:"Ù
زاÙ
ÙØ© ØªØ¯Ø±ÙØ¬ÙØ© Ù
ستÙ
رة Ù
ع إعادة ÙÙØ±Ø³Ø© ÙØ§Ù
ÙØ© Ø¹ÙØ¯ Ø§ÙØ·Ùب."},{q:"ÙÙ ÙØØªØ±Ù
ØµÙØ§ØÙات اÙÙ
صدرØ",a:"ÙØ¹Ù
â ÙØªÙ
Ø§ÙØÙØ§Ø¸ عÙÙ ØµÙØ§ØÙات Drive ÙSharePoint ÙNotion ÙConfluence ÙÙØª Ø§ÙØ§Ø³ØªØ¹ÙاÙ
."}]},"ai-revenue-intelligence":{name:"ÙÙÙÙ Ø°ÙØ§Ø¡ Ø§ÙØ¥Ùرادات Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",tagline:"Ø°ÙØ§Ø¡ Ø¥ÙØ±Ø§Ø¯Ø§Øª شاÙ
Ù â pipelineØ ØªÙÙØ¹Ø§ØªØ Ù
خاطر.",overview:"AI Revenue Intelligence ÙØÙÙ CRM ÙØ§ÙÙÙØªØ±Ø© ÙØ§Ø³ØªØ®Ø¯Ø§Ù
اÙÙ
ÙØªØ¬ باستÙ
رار ÙØªÙÙØ¹ Ø§ÙØ¥ÙØ±Ø§Ø¯Ø§ØªØ ØªØØ¯Ùد Ø§ÙØµÙÙØ§Øª اÙÙ
عرضة ÙÙØ®Ø·Ø±Ø ÙØ§ÙØªØ±Ø§Ø Ø£ÙØ¶Ù إجراء Ù
Ø¨ÙØ¹Ø§Øª تاÙÙ ÙÙÙ ØØ³Ø§Ø¨.",businessValue:["تØÙÙÙ pipeline اÙÙ
Ø¨ÙØ¹Ø§Øª Ù٠اÙÙÙØª Ø§ÙØÙÙÙÙ","تÙÙØ¹ Ø¥ÙØ±Ø§Ø¯Ø§Øª Ø£Ø³Ø¨ÙØ¹Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ","Ø¥ÙØ°Ø§Ø± Ù
Ø¨ÙØ± ÙÙ
خاطر Ø§ÙØµÙÙØ§Øª Ùchurn","ØªÙØ³ÙÙ
سÙÙÙ Ø§ÙØ¹Ù
ÙØ§Ø¡","Ø£ÙØ¶Ù إجراء تاÙÙ ÙÙÙ ØØ³Ø§Ø¨"],workflows:["تÙÙÙÙ
ØµØØ© pipeline","تÙÙØ¹ Ø¥ÙØ±Ø§Ø¯Ø§Øª Ø£Ø³Ø¨ÙØ¹Ù","تØÙÙÙ Ø§ÙØµÙÙØ§Øª Ø§ÙØ®Ø§Ø³Ø±Ø©","تØÙÙÙ cohort ÙØ³ÙÙÙ Ø§ÙØ¹Ù
ÙØ§Ø¡","Ù
ÙØªØ±Ø next-best-action"],roi:"ÙÙÙØ± 180 Ø£Ù٠دÙÙØ§Ø± سÙÙÙÙØ§ ÙÙ RevOps ÙÙÙØªØ 2.4 Ù
ÙÙÙ٠دÙÙØ§Ø± سÙÙÙÙØ§ Ù
Ù pipeline اÙÙ
ستعاد.",faq:[{q:"Ù
ا Ø£ÙØ¸Ù
Ø© CRM اÙÙ
دعÙÙ
Ø©Ø",a:"Salesforce ÙHubSpot ÙPipedrive ÙMicrosoft Dynamics 365 عبر Marketplace اÙÙ
ÙØµÙات."},{q:"Ù
ا Ø¯ÙØ© Ø§ÙØªÙÙØ¹Ø§ØªØ",a:"عادة 92-97% MAPE بعد ربع ÙØ§ØØ¯ Ù
Ù Ø§ÙØ§ØªØµØ§ÙØ Ù
Ø¶Ø¨ÙØ· عÙÙ ØªØ§Ø±ÙØ®Ù."},{q:"ÙÙ ÙÙ
ÙÙÙ Ø¯ÙØ¹ إجراءات Ø¥ÙÙ CRMØ",a:"ÙØ¹Ù
â Ù
ع Ù
ÙØ§ÙÙØ© Ø¨Ø´Ø±ÙØ© ÙØ³Ø¬Ù تدÙÙÙ ÙØ§Ù
Ù."}]},"ai-process-automation-auditor":{name:"Ù
دÙ٠أتÙ
تة Ø§ÙØ¹Ù
ÙÙØ§Øª Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",tagline:"ÙÙØ´Ù ÙÙ ÙØ±ØµØ© أتÙ
تة Ù
Ø®ÙÙØ© Ø¯Ø§Ø®Ù Ø´Ø±ÙØªÙ.",overview:"AI Process Automation Auditor ÙÙ
Ø³Ø Ø§ÙÙØ´Ø§Ø· اÙÙ
ØªÙØ±Ø± عبر Google Workspace ÙMicrosoft 365 ÙØ§Ùدردشة ÙERP ÙCRM ÙØ¥Ø¨Ø±Ø§Ø² ÙØ±Øµ Ø§ÙØ£ØªÙ
تة Ø§ÙØ£Ø¹ÙÙ Ø¹Ø§Ø¦Ø¯ÙØ§ ØØ³Ø¨ Ø§ÙØªÙÙÙØ± اÙÙ
تÙÙØ¹.",businessValue:["Ø§ÙØªØ´Ø§Ù Ù
ستÙ
ر ÙÙØ±Øµ Ø§ÙØ£ØªÙ
تة","Ø®Ø±ÙØ·Ø© ØØ±Ø§Ø±ÙØ© ÙÙÙ
ÙØ§Ù
اÙÙ
ØªÙØ±Ø±Ø© ØØ³Ø¨ اÙÙØ±ÙÙ","ÙØ§Ø¦Ù
Ø© أتÙ
تة Ù
رتبة ØØ³Ø¨ Ø§ÙØªÙÙÙØ±","ØªÙØ¯Ùر ROI ÙÙÙ workflow دÙÙ Ø¬ÙØ¯","Ù
Ø¹ÙØ§Ø± ÙÙØ§Ø¡Ø© عÙÙ Ù
ستÙ٠اÙÙ
ؤسسة"],workflows:["تÙÙÙØ¨ اÙÙØ´Ø§Ø· ÙØªØ¬Ù
ÙØ¹ اÙÙ
ÙØ§Ù
","ØªØ´Ø®ÙØµ workflows ØºÙØ± اÙÙØ¹Ø§ÙØ©","Backlog أتÙ
تة ÙØªØ±ØªÙب ROI","ØªÙØ±Ùر ØªØØ³ÙÙ ÙÙØ¥Ø¯Ø§Ø±Ø© Ø§ÙØ¹ÙÙØ§","تسÙÙÙ
Ø§ÙØªÙÙÙØ° Ø¥ÙÙ Ù
ÙØ¸ÙÙ AI"],roi:"ÙÙÙØ± 160 Ø£Ù٠دÙÙØ§Ø± سÙÙÙÙØ§ Ù٠تØÙÙÙ Ø§ÙØ¹Ù
ÙÙØ§Øª ÙÙÙØºÙ Ø£ÙØ«Ø± Ù
Ù 2 Ù
ÙÙÙ٠دÙÙØ§Ø± سÙÙÙÙØ§ Ù
Ù Ø§ÙØ¹Ù
٠اÙÙ
ØªÙØ±Ø±.",faq:[{q:"ÙÙ ÙØØªØ§Ø¬ Ø¥ÙÙ ÙØ±Ø§Ø¡Ø© Ù
Ø³ØªÙØ¯Ø§ØªÙاØ",a:"ÙÙØ±Ø£ metadata Ø³ÙØ± Ø§ÙØ¹Ù
Ù Ø§ÙØªØ±Ø§Ø¶ÙÙØ§. اÙÙØµÙ٠اÙÙØ§Ù
Ù ÙÙÙ
ØØªÙÙ Ø§Ø®ØªÙØ§Ø±Ù ÙÙÙ Ù
صدر."},{q:"ÙÙ ÙØªØµÙ بأ٠ERP Ø£Ù CRMØ",a:"ÙØ¹Ù
â SAP ÙOracle ÙNetSuite ÙSalesforce ÙHubSpot ÙDynamics 365 ÙÙÙØ§Ø¹Ø¯ Ø§ÙØ¨ÙØ§ÙØ§Øª."},{q:"Ù
ØªÙ ØªØ¸ÙØ± اÙÙØªØ§Ø¦Ø¬Ø",a:"Ø£ÙÙ backlog أتÙ
تة Ù
رتب Ø®ÙØ§Ù 7 Ø£ÙØ§Ù
Ù
٠ربط اÙÙ
صادر."}]}
811,"ai-employee-manager":{name:"Ù
Ø¯ÙØ± Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",tagline:"إدارة Ù
ÙØ¸ÙÙ
Ø© AI Workforce باÙÙØ§Ù
Ù â Ø§ÙØ£Ø¯Ø§Ø¡Ø Ø§ÙØªÙÙÙØ©Ø ÙØ§ÙØµÙØ§ØÙات.",overview:"AI Employee Manager ÙØªØªØ¨Ø¹ أداء ÙÙ Ù
ÙØ¸Ù AI ÙØ§Ø³ØªÙÙØ§Ù tokens ÙØ§ÙتÙÙÙØ©. ÙÙØµÙ Ø¨ØªØØ³Ù٠اÙÙÙ
Ø§Ø°Ø¬Ø ÙÙØªØ´Ù اÙÙ
ÙØ§Ù
اÙÙØ§Ø´ÙØ©Ø ÙØ¯Ùر Ø§ÙØµÙاØÙØ§ØªØ ÙÙÙØªØ¬ ØªÙØ±Ùر AI Workforce ÙØ§Ù
Ù ÙÙÙ CFO ÙØ§ÙÙ CTO.",businessValue:["Ù
Ø±Ø§ÙØ¨Ø© ØÙØ© ÙØ£Ø¯Ø§Ø¡ ÙÙ Ù
ÙØ¸Ù AI","تتبع tokens ÙØ§ÙتÙÙÙØ© ØØªÙ Ù
ستÙ٠اÙÙ
ÙÙ
Ø©","ØªÙØµÙات تÙÙØ§Ø¦ÙØ© ÙØªÙجÙ٠اÙÙÙ
اذج","ÙØ´Ù اÙÙ
ÙØ§Ù
اÙÙØ§Ø´ÙØ© ÙØ³Ùاسات auto-retry","اÙ
ØªØ«Ø§Ù ÙØµÙاØÙات ÙÙØ© Ø§ÙØ¹Ù
Ù Ø¨Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ"],workflows:["ÙÙØØ© أداء Ù
ÙØ¸ÙÙ AI","تØÙÙ٠استÙÙØ§Ù tokens ÙØ§ÙØªÙØ§ÙÙÙ","ØªÙØµÙØ© ØªØØ³Ù٠اÙÙÙ
اذج","ÙØ±Ø² ÙØ§Ø¦Ù
Ø© اÙÙ
ÙØ§Ù
اÙÙØ§Ø´ÙØ©","ØªÙØ±Ùر Ø§ÙØµÙاØÙات ÙØ§ÙتدÙÙÙ"],roi:"ÙÙÙØ± 120 Ø£Ù٠دÙÙØ§Ø± سÙÙÙÙØ§ ÙÙ AI Ops ÙÙÙØªØ 1.5 Ù
ÙÙÙ٠دÙÙØ§Ø± سÙÙÙÙØ§ Ù
Ù Ø®ÙØ¶ ØªÙØ§ÙÙÙ Ø§ÙØ§Ø³ØªØ¯ÙاÙ.",faq:[{q:"Ù
ا Ù
Ø²ÙØ¯Ù AI اÙÙ
تتبعÙÙØ",a:"Infernex API ÙOpenAI ÙAnthropic ÙGoogle Gemini ÙMistral ÙAzure AI ÙØ£Ù endpoint Ù
خصص."},{q:"ÙÙ ÙÙØ±Ø¶ ØØ¯Ùد Ø¥ÙÙØ§ÙØ",a:"ÙØ¹Ù
â ØØ³Ø¨ اÙÙ
ÙØ¸Ù ÙØ§ÙÙØ±ÙÙ ÙØ§ÙÙÙ
ÙØ°Ø¬Ø Ù
ع hard caps ÙØªÙبÙÙØ§Øª."},{q:"ÙÙ ÙÙ Ù
Ø·ÙÙØ¨Ø",a:"Ù
ÙØµÙ Ø¨Ù Ø¹ÙØ¯ تشغÙÙ Ø£ÙØ«Ø± Ù
Ù 3 Ù
ÙØ¸ÙÙ AI ÙÙ Ø§ÙØ¥Ùتاج."}]}}};for(const[e,t]of Object.entries(Xt)){const a=Ve[e]??(Ve[e]={}),o=a.marketplace??{},r=o.department??{},s=o.employeeItems??{};a.marketplace={...o,department:{...r,...t.department},employeeItems:{...s,...t.employeeItems}}}const qr={en:{how:{seoT:"How to create your company with Infernex AI â step by step",seoD:"The complete step-by-step process to hire AI employees and launch a fully operational company with Infernex AI, with real examples and cost savings figures.",badge:"Step-by-step guide",title:"How to build your company with Infernex AI",sub:"Six steps, one click each. From the first idea to a globally operating company run by autonomous AI employees.",problemTitle:"The problem",problemText:"Companies lose millions every year because repetitive knowledge work is still done manually.",solutionTitle:"The solution",solutionText:"Infernex deploys autonomous AI employees that perform this work 24/7 across every department.",cta:"Start now â free",cta2:"Back to home",stepsT:"The 6 steps of creation",s1t:"Describe your idea",s1d:"Type one paragraph: what you sell, to whom, in which country. The Master AI Agent turns it into a full business blueprint (positioning, offer, pricing, ICP, KPIs).",s1e:'Example: "A B2B SaaS for dental clinics in Switzerland." â blueprint generated in under 2 minutes.',s2t:"Validate the legal & financial structure",s2d:"Infernex selects the legal form, the applicable regulations (VAT, GDPR, nLPD, ISO 27001) and the starting budget. Every blocking item must be acknowledged by you before deployment.",s2e:"Example: Swiss Sà rl, VAT + nLPD + GDPR, CHF 20,000 seed budget â validated in one click.",s3t:"Hire your AI employees",s3d:"Pick your team from the marketplace: Sales, Support, Marketing, Finance, Research, Legal, and more. Each agent is operational immediately, 24/7, with no onboarding.",s3e:"Example: a 5-agent team replaces the workload of 8 junior full-time hires.",s4t:"Connect your tools",s4d:"93 connectors available (CRM, email, accounting, helpdesk, cloud, e-commerce). Your AI employees act directly inside your existing stack.",s4e:"Example: Stripe + HubSpot + Gmail connected in under 10 minutes, credentials encrypted AES-256.",s5t:"Deploy the company",s5d:"Brand, domain, website, cloud infrastructure, payments and operations are provisioned automatically. Every irreversible action stays behind the human approval engine.",s5e:"Example: domain + site + payment flow live in 24 hours instead of 6 weeks with an agency.",s6t:"Scale globally",s6d:"Multi-country, multi-currency, multi-language. The Master AI Agent forecasts cash, risk, sales and hiring needs, then delegates the work to your agents.",s6e:"Example: opening a second market (Germany) with legal + tax + localisation handled automatically.",saveT:"What it costs you â and what you save",saveS:"Comparison of a standard human team versus the same functions run by Infernex AI employees. Monthly figures, per function.",colRole:"Function",colHuman:"Human team / month",colIx:"Infernex / month",colSave:"Savings",r1:"Sales & prospecting",r2:"Customer support 24/7",r3:"Marketing & content",r4:"Finance & invoicing",r5:"Research & ops",totalRow:"Total",note:"Human costs are based on average loaded European salaries (junior to mid-level). Infernex costs include the subscription and inference routed to the cheapest engine that meets your SLA â up to 95% cheaper than calling frontier models directly.",ctaT:"Ready to launch your company?",ctaS:"Create your first AI employee for free, then deploy your company in one click.",ctaB:"Create my company"}},fr:{how:{seoT:"Comment créer votre entreprise avec Infernex AI â étape par étape",seoD:"Le processus complet, étape par étape, pour recruter des employés IA et lancer une entreprise pleinement opérationnelle avec Infernex AI, avec exemples réels et chiffres d'économies.",badge:"Guide étape par étape",title:"Comment créer votre entreprise avec Infernex AI",sub:"Six étapes, un clic chacune. De la première idée à une entreprise opérant à l'échelle mondiale, pilotée par des employés IA autonomes.",problemTitle:"Le problème",problemText:"Les entreprises perdent des millions chaque année parce que le travail répétitif de connaissance est encore fait manuellement.",solutionTitle:"La solution",solutionText:"Infernex déploie des employés IA autonomes qui effectuent ce travail 24/7 dans chaque département.",cta:"Commencer â gratuit",cta2:"Retour à l'accueil",stepsT:"Les 6 étapes de la création",s1t:"Décrivez votre idée",s1d:"Ãcrivez un paragraphe : ce que vous vendez, à qui, dans quel pays. L'Agent IA Maître le transforme en business blueprint complet (positionnement, offre, prix, ICP, KPI).",s1e
811:"Exemple : « Un SaaS B2B pour les cabinets dentaires en Suisse. » â blueprint généré en moins de 2 minutes.",s2t:"Validez la structure juridique et financière",s2d:"Infernex sélectionne la forme juridique, les réglementations applicables (TVA, RGPD, nLPD, ISO 27001) et le budget de départ. Chaque point bloquant doit être acquitté par vous avant le déploiement.",s2e:"Exemple : Sà rl suisse, TVA + nLPD + RGPD, budget de départ CHF 20'000 â validé en un clic.",s3t:"Recrutez vos employés IA",s3d:"Choisissez votre équipe dans la marketplace : Ventes, Support, Marketing, Finance, Recherche, Juridique, et plus. Chaque agent est opérationnel immédiatement, 24/7, sans onboarding.",s3e:"Exemple : une équipe de 5 agents remplace la charge de travail de 8 juniors à temps plein.",s4t:"Connectez vos outils",s4d:"93 connecteurs disponibles (CRM, email, comptabilité, helpdesk, cloud, e-commerce). Vos employés IA agissent directement dans votre stack existante.",s4e:"Exemple : Stripe + HubSpot + Gmail connectés en moins de 10 minutes, identifiants chiffrés AES-256.",s5t:"Déployez l'entreprise",s5d:"Marque, domaine, site web, infrastructure cloud, paiements et opérations sont provisionnés automatiquement. Toute action irréversible reste soumise au moteur d'approbation humaine.",s5e:"Exemple : domaine + site + flux de paiement en ligne en 24 heures au lieu de 6 semaines avec une agence.",s6t:"Passez à l'échelle mondiale",s6d:"Multi-pays, multi-devises, multi-langues. L'Agent IA Maître prévoit la trésorerie, les risques, les ventes et les besoins de recrutement, puis délègue le travail à vos agents.",s6e:"Exemple : ouverture d'un second marché (Allemagne) avec juridique + fiscalité + localisation gérés automatiquement.",saveT:"Ce que ça vous coûte â et ce que vous économisez",saveS:"Comparaison d'une équipe humaine standard face aux mêmes fonctions assurées par des employés IA Infernex. Chiffres mensuels, par fonction.",colRole:"Fonction",colHuman:"Ãquipe humaine / mois",colIx:"Infernex / mois",colSave:"Ãconomie",r1:"Ventes & prospection",r2:"Support client 24/7",r3:"Marketing & contenu",r4:"Finance & facturation",r5:"Recherche & opérations",totalRow:"Total",note:"Les coûts humains se basent sur des salaires chargés européens moyens (junior à confirmé). Les coûts Infernex incluent l'abonnement et l'inférence routée vers le moteur le moins cher respectant votre SLA â jusqu'à 95 % moins cher qu'un appel direct aux modèles frontier.",ctaT:"Prêt à lancer votre entreprise ?",ctaS:"Créez votre premier employé IA gratuitement, puis déployez votre entreprise en un clic.",ctaB:"Créer mon entreprise"}},es:{how:{seoT:"Cómo crear tu empresa con Infernex AI â paso a paso",seoD:"El proceso completo, paso a paso, para contratar empleados IA y lanzar una empresa plenamente operativa con Infernex AI, con ejemplos reales y cifras de ahorro.",badge:"GuÃa paso a paso",title:"Cómo crear tu empresa con Infernex AI",sub:"Seis pasos, un clic cada uno. De la primera idea a una empresa que opera a escala global, dirigida por empleados IA autónomos.",problemTitle:"El problema",problemText:"Las empresas pierden millones cada año porque el trabajo de conocimiento repetitivo aún se hace manualmente.",solutionTitle:"La solución",solutionText:"Infernex despliega empleados IA autónomos que realizan este trabajo 24/7 en cada departamento.",cta:"Empezar â gratis",cta2:"Volver al inicio",stepsT:"Los 6 pasos de la creación",s1t:"Describe tu idea",s1d:"Escribe un párrafo: qué vendes, a quién y en qué paÃs. El Agente Maestro de IA lo convierte en un blueprint de negocio completo (posicionamiento, oferta, precios, ICP, KPI).",s1e
811:"Ejemplo: «Un SaaS B2B para clÃnicas dentales en Suiza.» â blueprint generado en menos de 2 minutos.",s2t:"Valida la estructura legal y financiera",s2d:"Infernex selecciona la forma jurÃdica, las regulaciones aplicables (IVA, RGPD, nLPD, ISO 27001) y el presupuesto inicial. Cada punto bloqueante debe ser aceptado por ti antes del despliegue.",s2e:"Ejemplo: Sà rl suiza, IVA + nLPD + RGPD, presupuesto inicial de CHF 20 000 â validado en un clic.",s3t:"Contrata a tus empleados IA",s3d:"Elige tu equipo en el marketplace: Ventas, Soporte, Marketing, Finanzas, Investigación, Legal y más. Cada agente es operativo de inmediato, 24/7, sin onboarding.",s3e:"Ejemplo: un equipo de 5 agentes sustituye la carga de trabajo de 8 juniors a tiempo completo.",s4t:"Conecta tus herramientas",s4d:"93 conectores disponibles (CRM, email, contabilidad, helpdesk, cloud, e-commerce). Tus empleados IA actúan directamente en tu stack existente.",s4e:"Ejemplo: Stripe + HubSpot + Gmail conectados en menos de 10 minutos, credenciales cifradas AES-256.",s5t:"Despliega la empresa",s5d:"Marca, dominio, sitio web, infraestructura cloud, pagos y operaciones se aprovisionan automáticamente. Toda acción irreversible pasa por el motor de aprobación humana.",s5e:"Ejemplo: dominio + web + flujo de pago en lÃnea en 24 horas en lugar de 6 semanas con una agencia.",s6t:"Escala globalmente",s6d:"MultipaÃs, multidivisa, multiidioma. El Agente Maestro de IA prevé tesorerÃa, riesgos, ventas y necesidades de contratación, y delega el trabajo en tus agentes.",s6e:"Ejemplo: apertura de un segundo mercado (Alemania) con legal + fiscalidad + localización gestionados automáticamente.",saveT:"Lo que te cuesta â y lo que ahorras",saveS:"Comparación de un equipo humano estándar frente a las mismas funciones ejecutadas por empleados IA de Infernex. Cifras mensuales, por función.",colRole:"Función",colHuman:"Equipo humano / mes",colIx:"Infernex / mes",colSave:"Ahorro",r1:"Ventas y prospección",r2:"Soporte al cliente 24/7",r3:"Marketing y contenido",r4:"Finanzas y facturación",r5:"Investigación y operaciones",totalRow:"Total",note:"Los costes humanos se basan en salarios europeos medios con cargas (junior a intermedio). Los costes de Infernex incluyen la suscripción y la inferencia enrutada al motor más barato que cumpla tu SLA â hasta un 95 % más barato que llamar directamente a los modelos frontier.",ctaT:"¿Listo para lanzar tu empresa?",ctaS:"Crea tu primer empleado IA gratis y despliega tu empresa en un clic.",ctaB:"Crear mi empresa"}},de:{how:{seoT:"So gründen Sie Ihr Unternehmen mit Infernex AI â Schritt für Schritt",seoD:"Der komplette Schritt-für-Schritt-Prozess, um KI-Mitarbeiter einzustellen und mit Infernex AI ein voll funktionsfähiges Unternehmen zu starten â mit echten Beispielen und Einsparzahlen.",badge:"Schritt-für-Schritt-Anleitung",title:"So bauen Sie Ihr Unternehmen mit Infernex AI",sub:"Sechs Schritte, je ein Klick. Von der ersten Idee zu einem global agierenden Unternehmen, geführt von autonomen KI-Mitarbeitern.",problemTitle:"Das Problem",problemText:"Unternehmen verlieren jedes Jahr Millionen, weil repetitive Wissensarbeit noch manuell erledigt wird.",solutionTitle:"Die Lösung",solutionText:"Infernex setzt autonome KI-Mitarbeiter ein, die diese Arbeit 24/7 in jeder Abteilung erledigen.",cta:"Jetzt starten â kostenlos",cta2:"Zurück zur Startseite",stepsT:"Die 6 Schritte der Gründung",s1t:"Beschreiben Sie Ihre Idee",s1d:"Schreiben Sie einen Absatz: was Sie verkaufen, an wen, in welchem Land. Der Master-KI-Agent macht daraus einen vollständigen Business-Blueprint (Positionierung, Angebot, Preise, ICP, KPIs).",s1e:"Beispiel: âEin B2B-SaaS für Zahnarztpraxen in der Schweiz.â â Blueprint in unter 2 Minuten erstellt.",s2t:"Rechts- und Finanzstruktur validieren",s2d:"Infernex wählt Rechtsform, geltende Regulierungen (MwSt., DSGVO, nDSG, ISO 27001) und Startbudget. Jeder blockierende Punkt muss von Ihnen bestätigt werden, bevor deployt wird.",s2e:"Beispiel: Schweizer GmbH, MwSt. + nDSG + DSGVO, Startbudget CHF 20'000 â mit einem Klick bestätigt.",s3t:"Stellen Sie Ihre KI-Mitarbeiter ein",s3d:"Wählen Sie Ihr Team im Marketplace: Vertrieb, Support, Marketing, Finanzen, Research, Legal und mehr. Jeder Agent ist sofort einsatzbereit, 24/7, ohne Onboarding.",s3e:"Beispiel: Ein Team aus 5 Agenten ersetzt die Arbeitslast von 8 Vollzeit-Junioren.",s4t:"Verbinden Sie Ihre Tools",s4d:"93 Konnektoren verfügbar (CRM, E-Mail, Buchhaltung, Helpdesk, Cloud, E-Commerce). Ihre KI-Mitarbeiter arbeiten direkt in Ihrem bestehenden Stack.",s4e:"Beispiel: Stripe + HubSpot + Gmail in unter 10 Minuten verbunden, Zugangsdaten AES-256-verschlüsselt.",s5t:"Unternehmen deployen",s5d:"Marke, Domain, Website, Cloud-Infrastruktur, Zahlungen und Betrieb werden automatisch bereitgestellt. Jede irreversible Aktion bleibt hinter der menschlichen Freigabe.",s5e:"Beispiel: Domain + Website
811+ Zahlungsfluss in 24 Stunden live statt 6 Wochen mit einer Agentur.",s6t:"Global skalieren",s6d:"Multi-Country, Multi-Currency, Multi-Language. Der Master-KI-Agent prognostiziert Liquidität, Risiken, Umsatz und Personalbedarf und delegiert die Arbeit an Ihre Agenten.",s6e:"Beispiel: Eröffnung eines zweiten Markts (Deutschland) mit automatisch geregelten Rechts-, Steuer- und Lokalisierungsthemen.",saveT:"Was es kostet â und was Sie sparen",saveS:"Vergleich eines klassischen Teams mit denselben Funktionen durch Infernex-KI-Mitarbeiter. Monatliche Zahlen pro Funktion.",colRole:"Funktion",colHuman:"Menschliches Team / Monat",colIx:"Infernex / Monat",colSave:"Ersparnis",r1:"Vertrieb & Akquise",r2:"Kundensupport 24/7",r3:"Marketing & Content",r4:"Finanzen & Rechnungsstellung",r5:"Research & Operations",totalRow:"Gesamt",note:"Die Personalkosten basieren auf durchschnittlichen europäischen Vollkosten-Gehältern (Junior bis Mid-Level). Die Infernex-Kosten umfassen Abo und Inferenz, geroutet auf die günstigste Engine, die Ihr SLA erfüllt â bis zu 95 % günstiger als direkte Aufrufe von Frontier-Modellen.",ctaT:"Bereit, Ihr Unternehmen zu starten?",ctaS:"Erstellen Sie Ihren ersten KI-Mitarbeiter kostenlos und deployen Sie Ihr Unternehmen mit einem Klick.",ctaB:"Mein Unternehmen erstellen"}},pt:{how:{seoT:"Como criar a sua empresa com a Infernex AI â passo a passo",seoD:"O processo completo, passo a passo, para contratar funcionários IA e lançar uma empresa totalmente operacional com a Infernex AI, com exemplos reais e números de poupança.",badge:"Guia passo a passo",title:"Como criar a sua empresa com a Infernex AI",sub:"Seis passos, um clique cada. De uma primeira ideia a uma empresa a operar à escala global, gerida por funcionários IA autónomos.",problemTitle:"O problema",problemText:"As empresas perdem milhões todos os anos porque o trabalho repetitivo de conhecimento ainda é feito manualmente.",solutionTitle:"A solução",solutionText:"A Infernex implementa funcionários IA autónomos que realizam este trabalho 24/7 em todos os departamentos.",cta:"Começar â grátis",cta2:"Voltar ao inÃcio",stepsT:"Os 6 passos da criação",s1t:"Descreva a sua ideia",s1d:"Escreva um parágrafo: o que vende, a quem e em que paÃs. O Agente Mestre de IA transforma isso num blueprint de negócio completo (posicionamento, oferta, preços, ICP, KPIs).",s1e:"Exemplo: «Um SaaS B2B para clÃnicas dentárias na SuÃça.» â blueprint gerado em menos de 2 minutos.",s2t:"Valide a estrutura jurÃdica e financeira",s2d:"A Infernex seleciona a forma jurÃdica, as regulamentações aplicáveis (IVA, RGPD, nLPD, ISO 27001) e o orçamento inicial. Cada ponto bloqueador deve ser aceite por si antes do deployment.",s2e:"Exemplo: Sà rl suÃça, IVA + nLPD + RGPD, orçamento inicial de CHF 20 000 â validado num clique.",s3t:"Contrate os seus funcionários IA",s3d:"Escolha a sua equipa no marketplace: Vendas, Suporte, Marketing, Finanças, Research, JurÃdico e mais. Cada agente fica operacional de imediato, 24/7, sem onboarding.",s3e:"Exemplo: uma equipa de 5 agentes substitui a carga de trabalho de 8 juniores a tempo inteiro.",s4t:"Ligue as suas ferramentas",s4d:"93 conectores disponÃveis (CRM, email, contabilidade, helpdesk, cloud, e-commerce). Os seus funcionários IA atuam diretamente na sua stack existente.",s4e:"Exemplo: Stripe + HubSpot + Gmail ligados em menos de 10 minutos, credenciais cifradas AES-256.",s5t:"Faça o deployment da empresa",s5d:"Marca, domÃnio, website, infraestrutura cloud, pagamentos e operações são provisionados automaticamente. Toda ação irreversÃvel continua sujeita à aprovação humana.",s5e:"Exemplo: domÃnio + site + fluxo de pagamento online em 24 horas em vez de 6 semanas com uma agência.",s6t:"Escale globalmente",s6d:"MultipaÃs, multimoeda, multilÃngue. O Agente Mestre de IA prevê tesouraria, riscos, vendas e necessidades de contratação, e delega o trabalho aos seus agentes.",s6e:"Exemplo: abertura de um segundo mercado (Alemanha) com jurÃdico + fiscalidade + localização tratados automaticamente.",s
811aveT:"Quanto custa â e quanto poupa",saveS:"Comparação de uma equipa humana padrão com as mesmas funções executadas por funcionários IA da Infernex. Valores mensais, por função.",colRole:"Função",colHuman:"Equipa humana / mês",colIx:"Infernex / mês",colSave:"Poupança",r1:"Vendas e prospeção",r2:"Suporte ao cliente 24/7",r3:"Marketing e conteúdo",r4:"Finanças e faturação",r5:"Research e operações",totalRow:"Total",note:"Os custos humanos baseiam-se em salários europeus médios com encargos (júnior a intermédio). Os custos Infernex incluem a subscrição e a inferência encaminhada para o motor mais barato que cumpre o seu SLA â até 95 % mais barato do que chamar diretamente modelos frontier.",ctaT:"Pronto para lançar a sua empresa?",ctaS:"Crie o seu primeiro funcionário IA grátis e faça o deployment da empresa num clique.",ctaB:"Criar a minha empresa"}},zh:{how:{seoT:"å¦ä½ç¨ Infernex AI å建æ¨çå
¬å¸ â 忥æå",seoD:"宿´ç忥æµç¨ï¼éä½£ AI åå·¥å¹¶ç¨ Infernex AI å¯å¨ä¸å®¶å
¨é¢è¿è¥çå
¬å¸ï¼éç宿¡ä¾ä¸ææ¬èçæ°æ®ã",badge:"忥æå",title:"å¦ä½ç¨ Infernex AI å建æ¨çå
¬å¸",sub:"å
个æ¥éª¤ï¼æ¯æ¥ä¸é®ã仿åçæ³æ³å°ç±èªä¸» AI åå·¥è¿è¥çå
¨çåå
¬å¸ã",problemTitle:"é®é¢",problemText:"ä¼ä¸æ¯å¹´æå¤±æ°ç¾ä¸ï¼å 为é夿§çç¥è¯å·¥ä½ä»é äººå·¥å®æã",solutionTitle:"è§£å³æ¹æ¡",solutionText:"Infernex é¨ç½²èªä¸» AI åå·¥ï¼å¨å个é¨é¨ 7Ã24 å°æ¶æ§è¡è¿äºå·¥ä½ã",cta:"ç«å³å¼å§ â å
è´¹",cta2:"è¿åé¦é¡µ",stepsT:"å建å
¬å¸ç 6 个æ¥éª¤",s1t:"æè¿°æ¨çæ³æ³",s1d:"å䏿®µè¯ï¼æ¨åä»ä¹ãåç»è°ãå¨åªä¸ªå½å®¶ãä¸»æ§ AI æºè½ä½ä¼å°å
¶è½¬åä¸ºå®æ´çåä¸èå¾ï¼å®ä½ã产åãå®ä»·ãICPãKPIï¼ã",s1e:"示ä¾ï¼ãé¢åç士çç§è¯æç B2B SaaSããâ èå¾ 2 åéå
çæã",s2t:"确认æ³å¾ä¸è´¢å¡æ¶æ",s2d:"Infernex éæ©æ³å¾å½¢å¼ãéç¨æ³è§ï¼å¢å¼ç¨ãGDPRãnLPDãISO 27001ï¼åå¯å¨é¢ç®ãæ¯ä¸ªé»æé¡¹å¿
é¡»ç±æ¨ç¡®è®¤åæè½é¨ç½²ã",s2e:"示ä¾ï¼ç士æé责任å
¬å¸ï¼å¢å¼ç¨ + nLPD + GDPRï¼å¯å¨é¢ç® 2 ä¸çé â ä¸é®ç¡®è®¤ã",s3t:"éä½£æ¨ç AI åå·¥",s3d:"å¨å¸åºä¸æéå¢éï¼éå®ã客æãå¸åºãè´¢å¡ãç ç©¶ãæ³å¡çãæ¯ä¸ªæºè½ä½ç«å³ä¸å²ï¼å
¨å¤©å 24/7ï¼æ éå¹è®ã",s3e:"示ä¾ï¼5 个æºè½ä½å¢é坿¿ä»£ 8 åå
¨èå级åå·¥çå·¥ä½éã",s4t:"è¿æ¥æ¨çå·¥å
·",s4d:"93 个å¯ç¨è¿æ¥å¨ï¼CRMãé®ä»¶ãè´¢å¡ã客æãäºãçµåï¼ãæ¨ç AI åå·¥ç´æ¥å¨ç°æææ¯æ 䏿§è¡æä½ã",s4e:"示ä¾ï¼Stripe + HubSpot + Gmail 10 åéå
å®æè¿æ¥ï¼åæ® AES-256 å å¯ã",s5t:"é¨ç½²å
¬å¸",s5d:"åçãååãç½ç«ãäºåºç¡è®¾æ½ãæ¯ä»ä¸è¿è¥åèªå¨å¼éãææä¸å¯éæä½ä»é人工审æ¹ã",s5e:"示ä¾ï¼åå + ç½ç« + æ¯ä»æµç¨ 24 å°æ¶ä¸çº¿ï¼èéä¸ä»£çæºæåä½ç 6 å¨ã",s6t:"å
¨çæ©å¼ ",s6d:"å¤å½å®¶ãå¤è´§å¸ãå¤è¯è¨ãä¸»æ§ AI æºè½ä½é¢æµç°éæµãé£é©ãéå®åæèéæ±ï¼å¹¶å°å·¥ä½åæ´¾ç»æ¨çæºè½ä½ã",s6e:"示ä¾ï¼å¼æç¬¬äºä¸ªå¸åºï¼å¾·å½ï¼ï¼æ³å¾ãç¨å¡ä¸æ¬å°åèªå¨å¤çã",saveT:"ææ¬ä¸èç",saveS:"æ å人åå¢éä¸åçèè½ç± Infernex AI åå·¥æ§è¡ç对æ¯ãæèè½çæåº¦æ°æ®ã",colRole:"èè½",colHuman:"人åå¢é / æ",colIx:"Infernex / æ",colSave:"èç",r1:"éå®ä¸å¼æ",r2:"7Ã24 å®¢æ·æ¯æ",r3:"å¸åºä¸å
容",r4:"è´¢å¡ä¸å¼ç¥¨",r5:"ç ç©¶ä¸è¿è¥",totalRow:"å计",note:"äººåææ¬åºäºæ¬§æ´²å¹³åå«ç¨èªèµï¼å级è³ä¸çº§ï¼ãInfernex ææ¬å
å«è®¢é
è´¹åè·¯ç±è³æ»¡è¶³ SLA çæä½ææ¬å¼æçæ¨çè´¹ç¨ â æ¯ç´æ¥è°ç¨å沿模åæå¤ä¾¿å® 95%ã",ctaT:"åå¤å¥½å¯å¨æ¨çå
¬å¸äºåï¼",ctaS:"å
è´¹å建第ä¸ä¸ª AI åå·¥ï¼ç¶åä¸é®é¨ç½²æ¨çå
¬å¸ã",ctaB:"å建æçå
¬å¸"}},ja:{how:{seoT:"Infernex AIã§ä¼ç¤¾ãè¨ç«ããæ¹æ³ â ã¹ããããã¤ã¹ããã",seoD:"AI徿¥å¡ãéããInfernex AIã§å®å
¨ã«ç¨¼åããä¼ç¤¾ãç«ã¡ä¸ããã¾ã§ã®å
¨ã
811¹ãããããå
·ä½ä¾ã¨ã³ã¹ãåæ¸ã®æ°å¤ã¨ã¨ãã«è§£èª¬ãã¾ãã",badge:"ã¹ããããã¤ã¹ãããã»ã¬ã¤ã",title:"Infernex AIã§ä¼ç¤¾ãè¨ç«ããæ¹æ³",sub:"6ã¹ããããããããã¯ã³ã¯ãªãã¯ãæåã®ã¢ã¤ãã¢ãããèªå¾åAI徿¥å¡ãéå¶ããã°ãã¼ãã«ä¼æ¥ã¸ã",problemTitle:"åé¡",problemText:"伿¥ã¯æ¯å¹´æ°ç¾ä¸åä½ã®æå¤±ã被ã£ã¦ãã¾ãããªããªããç¹°ãè¿ãã®ç¥è使¥ã¯æªã ã«æä½æ¥ã§è¡ããã¦ããããã§ãã",solutionTitle:"解決ç",solutionText:"Infernexã¯èªå¾åAI徿¥å¡ãå±éããããããé¨éã§24æé365æ¥ãã®ä½æ¥ãå®è¡ãã¾ãã",cta:"ä»ããéå§ â ç¡æ",cta2:"ãã¼ã ã«æ»ã",stepsT:"è¨ç«ã¾ã§ã®6ã¹ããã",s1t:"ã¢ã¤ãã¢ãæ¸ã",s1d:"䏿®µè½ã ãè¨å
¥ãã¾ãï¼ä½ãã誰ã«ãã©ã®å½ã§å£²ãã®ãããã¹ã¿ã¼AIã¨ã¼ã¸ã§ã³ããå®å
¨ãªãã¸ãã¹ãã«ã¼ããªã³ãï¼ãã¸ã·ã§ãã³ã°ããªãã¡ã¼ãä¾¡æ ¼ãICPãKPIï¼ã«å¤æãã¾ãã",s1e:"ä¾ï¼ãã¹ã¤ã¹ã®æ¯ç§å»é¢åãB2B SaaSãâ 2å以å
ã«ãã«ã¼ããªã³ãçæã",s2t:"æ³åã»è²¡åã¹ãã©ã¯ãã£ã¼ãæ¿èª",s2d:"Infernexãæ³äººå½¢æ
ãé©ç¨è¦å¶ï¼VATãGDPRãnLPDãISO 27001ï¼ãåæäºç®ãé¸å®ãã¾ããããããã³ã°é
ç®ã¯ãããã¤åã«ããªãã®æ¿èªãå¿
è¦ã§ãã",s2e:"ä¾ï¼ã¹ã¤ã¹ã®Sà rlãVAT + nLPD + GDPRãåæäºç®CHF 20,000 â ã¯ã³ã¯ãªãã¯ã§æ¿èªã",s3t:"AI徿¥å¡ãéç¨",s3d:"ãã¼ã±ãããã¬ã¤ã¹ã§ãã¼ã ã鏿ï¼å¶æ¥ããµãã¼ãããã¼ã±ã財åããªãµã¼ããæ³åãªã©ãåã¨ã¼ã¸ã§ã³ãã¯ãªã³ãã¼ãã£ã³ã°ä¸è¦ã§å³æ¥24/7稼åã",s3e:"ä¾ï¼5ã¨ã¼ã¸ã§ã³ãã®ãã¼ã ããã«ã¿ã¤ã ã®ã¸ã¥ãã¢8ååã®ä½æ¥éã代æ¿ã",s4t:"ãã¼ã«ãæ¥ç¶",s4d:"93ã®ã³ãã¯ã¿ï¼CRMãã¡ã¼ã«ãä¼è¨ããã«ããã¹ã¯ãã¯ã©ã¦ããECï¼ãAI徿¥å¡ã¯æ¢åã¹ã¿ãã¯å
ã§ç´æ¥ç¨¼åãã¾ãã",s4e:"ä¾ï¼Stripe + HubSpot + Gmailã10å以å
ã«æ¥ç¶ãèªè¨¼æ
å ±ã¯AES-256ã§æå·åã",s5t:"ä¼ç¤¾ããããã¤",s5d:"ãã©ã³ãããã¡ã¤ã³ãã¦ã§ããµã¤ããã¯ã©ã¦ãåºç¤ã決æ¸ããªãã¬ã¼ã·ã§ã³ãèªåæ§ç¯ãä¸å¯éãªæä½ã¯äººéã®æ¿èªã¨ã³ã¸ã³ãçµç±ãã¾ãã",s5e:"ä¾ï¼ãã¡ã¤ã³ + ãµã¤ã + 決æ¸ããã¼ã代çåºã®6é±éã§ã¯ãªã24æéã§å
¬éã",s6t:"ã°ãã¼ãã«ã«ã¹ã±ã¼ã«",s6d:"ãã«ãå½ã»ãã«ãé貨ã»ãã«ãè¨èªããã¹ã¿ã¼AIã¨ã¼ã¸ã§ã³ããè³éç¹°ãããªã¹ã¯ã売ä¸ãæ¡ç¨ãã¼ãºãäºæ¸¬ããã¨ã¼ã¸ã§ã³ãã«æ¥åãå§ä»»ãã¾ãã",s6e:"ä¾ï¼ç¬¬2å¸å ´ï¼ãã¤ãï¼ã¸ã®é²åºã§ãæ³åã»ç¨åã»ãã¼ã«ã©ã¤ãºãèªåå¦çã",saveT:"ã³ã¹ãã¨åæ¸å¹æ",saveS:"æ¨æºçãªäººçãã¼ã ã¨ãåãæ©è½ãInfernexã®AI徿¥å¡ãæ
ãå ´åã®æ¯è¼ãæ©è½å¥ã®æé¡ã§ãã",colRole:"æ©è½",colHuman:"人çãã¼ã / æ",colIx:"Infernex / æ",colSave:"忏é¡",r1:"å¶æ¥ã»éæ",r2:"24/7ã«ã¹ã¿ãã¼ãµãã¼ã",r3:"ãã¼ã±ãã£ã³ã°ã»ã³ã³ãã³ã",r4:"財åã»è«æ±",r5:"ãªãµã¼ãã»ãªãã¬ã¼ã·ã§ã³",totalRow:"åè¨",note:"äººä»¶è²»ã¯æ¬§å·ã®å¹³åç·é¡äººä»¶è²»ï¼ã¸ã¥ãã¢ãããã«ï¼ã«åºã¥ãã¾ããInfernexã®ã³ã¹ãã¯ãµãã¹ã¯ãªãã·ã§ã³ã¨ãSLAãæºããæå®ã¨ã³ã¸ã³ã¸ã«ã¼ãã£ã³ã°ãããæ¨è«è²»ç¨ãå«ã¿ãããã³ãã£ã¢ã¢ãã«ã®ç´æ¥å¼ã³åºãããæå¤§95%å®ä¾¡ã§ãã",ctaT:"ä¼ç¤¾ãç«ã¡ä¸ããæºåã¯ã§ãã¾ãããï¼",ctaS:"æåã®AI徿¥å¡ãç¡æã§ä½æããã¯ã³ã¯ãªãã¯ã§ä¼ç¤¾ããããã¤ãã¾ãããã",ctaB:"ä¼ç¤¾ã使ãã"}},hi:{how:{seoT:"Infernex AI सॠà¤
पनॠà¤à¤à¤ªà¤¨à¥ à¤à¥à¤¸à¥ बनाà¤à¤ â à¤à¤°à¤£ दर à¤à¤°à¤£",seoD:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ नियà¥à¤à¥à¤¤ à¤à¤°à¤¨à¥ à¤à¤° Infernex AI à¤à¥ साथ पà¥à¤°à¥ तरह सà¤à¤à¤¾à¤²à¤¿à¤¤ à¤à¤à¤ªà¤¨à¥ शà¥à¤°à¥ à¤à¤°à¤¨à¥ à¤à¥ सà¤à¤ªà¥à¤°à¥à¤£ à¤à¤°à¤£-दर-à¤à¤°à¤£ पà¥à¤°à¤à¥à¤°à¤¿à¤¯à¤¾, वासà¥à¤¤à¤µà¤¿à¤ à¤à¤¦à¤¾à¤¹à¤°à¤£à¥à¤ à¤à¤° बà¤à¤¤ à¤à¥ à¤à¤à¤à¤¡à¤¼à¥à¤ à¤à¥ साथ।",badge:"à¤à¤°à¤£-दर-à¤à¤°à¤£ à¤à¤¾à¤à¤¡",title:"Infernex AI सॠà¤
पनॠà¤à¤à¤ªà¤¨à¥ à¤à¥à¤¸à¥ बनाà¤à¤",sub:"à¤à¤¹ à¤à¤°à¤£, हर à¤à¤°à¤£ à¤à¤ à¤à¥à¤²à¤¿à¤à¥¤ à¤à¤ विà¤à¤¾à¤° सॠलà¥à¤à¤° सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ दà¥à¤µà¤¾à¤°à¤¾ सà¤à¤à¤¾à¤²à¤¿à¤¤ वà¥à¤¶à¥à¤µà¤¿à¤ à¤à¤à¤ªà¤¨à¥ तà¤à¥¤",problemTitle:"समसà¥à¤¯à¤¾",problemText:"à¤à¤à¤ªà¤¨à¤¿à¤¯à¤¾à¤ हर साल लाà¤à¥à¤ डà¥à¤²à¤° à¤à¤à¤µà¤¾à¤¤à¥ हà¥à¤ à¤à¥à¤¯à¥à¤à¤à¤¿ दà¥à¤¹à¤°à¤¾à¤µ वाला à¤à¥à¤à¤¾à¤¨ à¤à¤¾à¤°à¥à¤¯ à¤
à¤à¥ à¤à¥ हाथ सॠà¤à¤¿à¤¯à¤¾ à¤à¤¾à¤¤à¤¾ हà¥à¥¤",
811solutionTitle:"समाधान",solutionText:"Infernex सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ à¤à¥ तà¥à¤¨à¤¾à¤¤ à¤à¤°à¤¤à¤¾ हॠà¤à¥ हर विà¤à¤¾à¤ मà¥à¤ यह à¤à¤¾à¤® 24/7 à¤à¤°à¤¤à¥ हà¥à¤à¥¤",cta:"à¤
à¤à¥ शà¥à¤°à¥ à¤à¤°à¥à¤ â निà¤à¤¶à¥à¤²à¥à¤",cta2:"हà¥à¤® पर लà¥à¤à¥à¤",stepsT:"निरà¥à¤®à¤¾à¤£ à¤à¥ 6 à¤à¤°à¤£",s1t:"à¤
पना विà¤à¤¾à¤° लिà¤à¥à¤",s1d:"à¤à¤ पà¥à¤°à¤¾à¤à¥à¤°à¤¾à¤« लिà¤à¥à¤: à¤à¤ª à¤à¥à¤¯à¤¾ बà¥à¤à¤¤à¥ हà¥à¤, à¤à¤¿à¤¸à¥, à¤à¤° à¤à¤¿à¤¸ दà¥à¤¶ मà¥à¤à¥¤ मासà¥à¤à¤° AI à¤à¤à¥à¤à¤ à¤à¤¸à¥ पà¥à¤°à¥à¤£ बिà¤à¤¼à¤¨à¥à¤¸ बà¥à¤²à¥à¤ªà¥à¤°à¤¿à¤à¤ (पà¥à¤à¤¼à¤¿à¤¶à¤¨à¤¿à¤à¤, à¤à¤«à¤°, पà¥à¤°à¤¾à¤à¤¸à¤¿à¤à¤, ICP, KPI) मà¥à¤ बदल दà¥à¤¤à¤¾ हà¥à¥¤",s1e:"à¤à¤¦à¤¾à¤¹à¤°à¤£: «सà¥à¤µà¤¿à¤à¥à¤à¤¼à¤°à¤²à¥à¤à¤¡ à¤à¥ डà¥à¤à¤à¤² à¤à¥à¤²à¥à¤¨à¤¿à¤à¥à¤ à¤à¥ लिठB2B SaaS» â 2 मिनठसॠà¤à¤® मà¥à¤ बà¥à¤²à¥à¤ªà¥à¤°à¤¿à¤à¤ तà¥à¤¯à¤¾à¤°à¥¤",s2t:"à¤à¤¾à¤¨à¥à¤¨à¥ à¤à¤° वितà¥à¤¤à¥à¤¯ ढाà¤à¤à¤¾ सà¥à¤µà¥à¤à¥à¤¤ à¤à¤°à¥à¤",s2d:"Infernex à¤à¤¾à¤¨à¥à¤¨à¥ सà¥à¤µà¤°à¥à¤ª, लाà¤à¥ नियम (VAT, GDPR, nLPD, ISO 27001) à¤à¤° पà¥à¤°à¤¾à¤°à¤à¤à
811¤¿à¤ बà¤à¤ à¤à¥à¤¨à¤¤à¤¾ हà¥à¥¤ हर à¤
वरà¥à¤§à¤ बिà¤à¤¦à¥ पर परिनियà¥à¤à¤¨ सॠपहलॠà¤à¤ªà¤à¥ सà¥à¤µà¥à¤à¥à¤¤à¤¿ à¤à¤¼à¤°à¥à¤°à¥ हà¥à¥¤",s2e:"à¤à¤¦à¤¾à¤¹à¤°à¤£: सà¥à¤µà¤¿à¤¸ Sà rl, VAT + nLPD + GDPR, पà¥à¤°à¤¾à¤°à¤à¤à¤¿à¤ बà¤à¤ CHF 20,000 â à¤à¤ à¤à¥à¤²à¤¿à¤ मà¥à¤ सà¥à¤µà¥à¤à¥à¤¤à¥¤",s3t:"à¤
पनॠAI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ नियà¥à¤à¥à¤¤ à¤à¤°à¥à¤",s3d:"मारà¥à¤à¥à¤à¤ªà¥à¤²à¥à¤¸ सॠà¤à¥à¤® à¤à¥à¤¨à¥à¤: सà¥à¤²à¥à¤¸, सपà¥à¤°à¥à¤, मारà¥à¤à¥à¤à¤¿à¤à¤, फाà¤à¤¨à¥à¤à¤¸, रिसरà¥à¤, लà¥à¤à¤² à¤à¤° à¤
धिà¤à¥¤ हर à¤à¤à¥à¤à¤ बिना à¤à¤¨à¤¬à¥à¤°à¥à¤¡à¤¿à¤à¤ à¤à¥ तà¥à¤°à¤à¤¤ 24/7 à¤à¤¾à¤® शà¥à¤°à¥ à¤à¤°à¤¤à¤¾ हà¥à¥¤",s3e:"à¤à¤¦à¤¾à¤¹à¤°à¤£: 5 à¤à¤à¥à¤à¤à¥à¤ à¤à¥ à¤à¥à¤® 8 फà¥à¤²-à¤à¤¾à¤à¤® à¤à¥à¤¨à¤¿à¤¯à¤° à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ à¤à¤¾ à¤à¤¾à¤°à¥à¤¯à¤à¤¾à¤° सà¤à¤à¤¾à¤²à¤¤à¥ हà¥à¥¤",s4t:"à¤
पनॠà¤à¥à¤² à¤à¥à¤¡à¤¼à¥à¤",s4d:"93 à¤à¤¨à¥à¤à¥à¤à¤° à¤à¤ªà¤²à¤¬à¥à¤§ (CRM, à¤à¤®à¥à¤², à¤
à¤à¤¾à¤à¤à¤à¤¿à¤à¤, हà¥à¤²à¥à¤ªà¤¡à¥à¤¸à¥à¤, à¤à¥à¤²à¤¾à¤à¤¡, à¤-à¤à¥à¤®à¤°à¥à¤¸)। à¤à¤ªà¤à¥ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¤ªà¤à¥ मà¥à¤à¥à¤¦à¤¾ सà¥à¤à¥à¤ मà¥à¤ सà¥à¤§à¥ à¤à¤¾à¤® à¤à¤°à¤¤à¥ हà¥à¤à¥¤",s4e:"à¤à¤¦à¤¾à¤¹à¤°à¤£: Stripe + HubSpot + Gmail 10 मिनठसॠà¤à¤® मà¥à¤ à¤à¥à¤¡à¤¼à¥, à¤à¥à¤°à¥à¤¡à¥à¤à¤¶à¤¿à¤¯à¤² AES-256 à¤à¤¨à¥à¤à¥à¤°à¤¿à¤ªà¥à¤à¥à¤¡à¥¤",s5t:"à¤à¤à¤ªà¤¨à¥ परिनियà¥à¤à¤¿à¤¤ à¤à¤°à¥à¤",s5d:"बà¥à¤°à¤¾à¤à¤¡, डà¥à¤®à¥à¤¨, वà¥à¤¬à¤¸à¤¾à¤à¤, à¤à¥à¤²à¤¾à¤à¤¡ à¤à¤à¤«à¥à¤°à¤¾à¤¸à¥à¤à¥à¤°à¤à¥à¤à¤°, à¤à¥à¤à¤¤à¤¾à¤¨ à¤à¤° सà¤à¤à¤¾à¤²à¤¨ सà¥à¤µà¤¤à¤ तà¥à¤¯à¤¾à¤° हà¥à¤¤à¥ हà¥à¤à¥¤ हर à¤
परिवरà¥à¤¤à¤¨à¥à¤¯ à¤à¤¾à¤°à¥à¤°à¤µà¤¾à¤ मानव à¤
नà¥à¤®à¥à¤¦à¤¨ à¤à¥ à¤
धà¥à¤¨ रहतॠहà¥à¥¤",s5e:"à¤à¤¦à¤¾à¤¹à¤°à¤£: डà¥à¤®à¥à¤¨ + साà¤à¤ + à¤à¥à¤à¤¤à¤¾à¤¨ पà¥à¤°à¤µà¤¾à¤¹ à¤à¤à¥à¤à¤¸à¥ à¤à¥ 6 सपà¥à¤¤à¤¾à¤¹ à¤à¥ बà¤à¤¾à¤¯ 24 à¤à¤à¤à¥ मà¥à¤ लाà¤à¤µà¥¤",s6t:"वà¥à¤¶à¥à¤µà¤¿à¤ सà¥à¤¤à¤° पर बढ़à¥à¤",s6d:"मलà¥à¤à¥-à¤à¤à¤à¥à¤°à¥, मलà¥à¤à¥-à¤à¤°à¥à¤à¤¸à¥, मलà¥à¤à¥-लà¥à¤à¤à¥à¤µà¥à¤à¥¤ मासà¥à¤à¤° AI à¤à¤à¥à¤à¤ नà¤à¤¦à¥, à¤à¥à¤à¤¿à¤®, बिà¤à¥à¤°à¥ à¤à¤° à¤à¤°à¥à¤¤à¥ à¤à¤µà¤¶à¥à¤¯à¤à¤¤à¤¾à¤à¤ à¤à¤¾ पà¥à¤°à¥à¤µà¤¾à¤¨à¥à¤®à¤¾à¤¨ लà¤à¤¾à¤à¤° à¤à¤¾à¤® à¤à¤ªà¤à¥ à¤à¤à¥à¤à¤à¥à¤ à¤à¥ सà¥à¤à¤ªà¤¤à¤¾ हà¥à¥¤",s6e:"à¤à¤¦à¤¾à¤¹à¤°à¤£: दà¥à¤¸à¤°à¤¾ बाà¤à¤¼à¤¾à¤° (à¤à¤°à¥à¤®à¤¨à¥) à¤à¥à¤²à¤¨à¤¾ â à¤à¤¾à¤¨à¥à¤¨à¥, à¤à¤° à¤à¤° सà¥à¤¥à¤¾à¤¨à¥à¤¯à¤à¤°à¤£ सà¥à¤µà¤¤à¤ सà¤à¤à¤¾à¤²à¥ à¤à¤à¥¤",saveT:"लाà¤à¤¤ â à¤à¤° à¤à¤ªà¤à¥ बà¤à¤¤",saveS:"मानठमानव à¤à¥à¤® बनाम वहॠà¤à¤¾à¤°à¥à¤¯ Infernex AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ दà¥à¤µà¤¾à¤°à¤¾à¥¤ पà¥à¤°à¤¤à¤¿ à¤à¤¾à¤°à¥à¤¯ मासिठà¤à¤à¤à¤¡à¤¼à¥à¥¤",colRole:"à¤à¤¾à¤°à¥à¤¯",colHuman:"मानव à¤à¥à¤® / माह",colIx:"Infernex / माह",colSave:"बà¤à¤¤",r1:"सà¥à¤²à¥à¤¸ à¤à¤° पà¥à¤°à¥à¤¸à¥à¤ªà¥à¤à¥à¤à¤¿à¤à¤",r2:"24/7 à¤à¥à¤°à¤¾à¤¹à¤ सहायता",r3:"मारà¥à¤à¥à¤à¤¿à¤à¤ à¤à¤° à¤à¤à¤à¥à¤à¤",r4:"फाà¤à¤¨à¥à¤à¤¸ à¤à¤° बिलिà¤à¤",r5:"रिसरà¥à¤ à¤à¤° à¤à¤ªà¤°à¥à¤¶à¤à¤¸",totalRow:"à¤à¥à¤²",note:"मानव लाà¤à¤¤ à¤à¤¸à¤¤ यà¥à¤°à¥à¤ªà¥à¤¯ à¤à¥à¤² वà¥à¤¤à¤¨ (à¤à¥à¤¨à¤¿à¤¯à¤° सॠमिड-लà¥à¤µà¤²) पर à¤à¤§à¤¾à¤°à¤¿à¤¤ हà¥à¥¤ Infernex लाà¤à¤¤ मà¥à¤ सबà¥à¤¸à¤à¥à¤°à¤¿à¤ªà¥à¤¶à¤¨ à¤à¤° à¤à¤ªà¤à¥ SLA à¤à¥ पà¥à¤°à¤¾ à¤à¤°à¤¨à¥ वालॠसबसॠससà¥à¤¤à¥ à¤à¤à¤à¤¨ पर रà¥à¤ à¤à¤¿à¤¯à¤¾ à¤à¤¯à¤¾ à¤à¤¨à¥à¤«à¤°à¥à¤à¤¸ शामिल हॠâ फà¥à¤°à¤à¤à¤¿à¤¯à¤° मà¥à¤¡à¤² à¤à¥ सà¥à¤§à¥ à¤à¥à¤² à¤à¤°à¤¨à¥ सॠ95% तठससà¥à¤¤à¤¾à¥¤",ctaT:"à¤
पनॠà¤à¤à¤ªà¤¨à¥ शà¥à¤°à¥ à¤à¤°à¤¨à¥ à¤à¥ लिठतà¥à¤¯à¤¾à¤° हà¥à¤?",ctaS:"à¤
पना पहला AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ मà¥à¤«à¤¼à¥à¤¤ बनाà¤à¤, फिर à¤à¤ à¤à¥à¤²à¤¿à¤ मà¥à¤ à¤à¤à¤ªà¤¨à¥ परिनियà¥à¤à¤¿à¤¤ à¤à¤°à¥à¤à¥¤",ctaB:"मà¥à¤°à¥ à¤à¤à¤ªà¤¨à¥ बनाà¤à¤"}},ar:{how:{seoT:"ÙÙÙ ØªÙØ´Ø¦ Ø´Ø±ÙØªÙ Ù
ع Infernex AI â Ø®Ø·ÙØ© Ø¨Ø®Ø·ÙØ©",seoD:"Ø§ÙØ¹Ù
ÙÙØ© اÙÙØ§Ù
ÙØ© Ø®Ø
811·ÙØ© Ø¨Ø®Ø·ÙØ© ÙØªÙظÙÙ Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù ÙØ¥Ø·ÙØ§Ù Ø´Ø±ÙØ© تعÙ
٠باÙÙØ§Ù
Ù Ù
ع Infernex AIØ Ù
ع Ø£Ù
Ø«ÙØ© ÙØ§ÙØ¹ÙØ© ÙØ£Ø±ÙاÙ
Ø§ÙØªÙÙÙØ±.",badge:"دÙÙÙ Ø®Ø·ÙØ© Ø¨Ø®Ø·ÙØ©",title:"ÙÙÙ ØªÙØ´Ø¦ Ø´Ø±ÙØªÙ Ù
ع Infernex AI",sub:"ست Ø®Ø·ÙØ§ØªØ ÙÙ Ø®Ø·ÙØ© بÙÙØ±Ø© ÙØ§ØØ¯Ø©. Ù
٠اÙÙÙØ±Ø© Ø§ÙØ£ÙÙ٠إÙÙ Ø´Ø±ÙØ© تعÙ
٠عÙÙ ÙØ·Ø§Ù عاÙÙ
Ù ÙØ¯ÙØ±ÙØ§ Ù
ÙØ¸ÙÙ Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù Ù
ستÙÙÙÙ.",problemTitle:"اÙÙ
Ø´ÙÙØ©",problemText:"تخسر Ø§ÙØ´Ø±Ùات Ù
ÙØ§ÙÙÙ Ø§ÙØ¯ÙÙØ§Ø±Ø§Øª Ù٠عاÙ
ÙØ£Ù Ø§ÙØ¹Ù
٠اÙÙ
عرÙ٠اÙÙ
ØªÙØ±Ø± ÙØ§ ÙØ²Ø§Ù ÙÙÙÙÙØ° ÙØ¯ÙÙÙØ§.",solutionTitle:"Ø§ÙØÙ",solutionText:"ØªÙØ´Ø± Infernex Ù
ÙØ¸ÙÙÙ Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù Ù
ستÙÙÙÙ ÙØ¤Ø¯ÙÙ ÙØ°Ø§ Ø§ÙØ¹Ù
٠عÙÙ Ù
دار Ø§ÙØ³Ø§Ø¹Ø© ÙÙ ÙÙ ÙØ³Ù
.",cta:"ابدأ Ø§ÙØ¢Ù â Ù
جاÙÙØ§",cta2:"Ø§ÙØ¹Ùدة Ø¥ÙÙ Ø§ÙØµÙØØ© Ø§ÙØ±Ø¦ÙØ³ÙØ©",stepsT:"Ø®Ø·ÙØ§Øª Ø§ÙØ¥Ùشاء Ø§ÙØ³Øª",s1t:"ص٠ÙÙØ±ØªÙ",s1d:"Ø§ÙØªØ¨ ÙÙØ±Ø© ÙØ§ØØ¯Ø©: Ù
اذا ØªØ¨ÙØ¹Ø ÙÙÙ
ÙØ ÙÙÙ Ø£Ù Ø¨ÙØ¯. ÙØÙÙÙÙØ§ اÙÙÙÙÙ Ø§ÙØ°ÙÙ Ø§ÙØ±Ø¦Ùس٠إÙÙ Ù
خطط أعÙ
Ø§Ù ÙØ§Ù
Ù (Ø§ÙØªÙ
ÙØ¶Ø¹Ø Ø§ÙØ¹Ø±Ø¶Ø Ø§ÙØªØ³Ø¹ÙØ±Ø ICPØ Ù
ؤشرات Ø§ÙØ£Ø¯Ø§Ø¡).",s1e:"Ù
ثاÙ: «Ù
ÙØµØ© SaaS ÙÙØ´Ø±Ùات Ù
ÙØ¬ÙØ© ÙØ¹Ùادات Ø§ÙØ£Ø³Ùا٠Ù٠سÙÙØ³Ø±Ø§.» â Ù
خطط Ø¬Ø§ÙØ² Ù٠أÙÙ Ù
٠دÙÙÙØªÙÙ.",s2t:"اعتÙ
د اÙÙÙÙ٠اÙÙØ§ÙÙÙÙ ÙØ§ÙÙ
اÙÙ",s2d:"ÙØ®ØªØ§Ø± Infernex Ø§ÙØ´Ù٠اÙÙØ§ÙÙÙÙ ÙØ§ÙÙÙØ§Ø¦Ø اÙÙ
Ø·Ø¨ÙØ© (Ø¶Ø±ÙØ¨Ø© اÙÙÙÙ
Ø© اÙÙ
Ø¶Ø§ÙØ©Ø GDPRØ nLPDØ ISO 27001) ÙÙ
ÙØ²Ø§ÙÙØ© Ø§ÙØ§ÙØ·ÙØ§Ù. ÙØ¬Ø¨ اعتÙ
اد ÙÙ Ø¨ÙØ¯ Ù
عÙÙ Ù
Ù ÙØ¨ÙÙ ÙØ¨Ù اÙÙØ´Ø±.",s2e:"Ù
ثاÙ: Ø´Ø±ÙØ© سÙÙØ³Ø±ÙØ© Sà rlØ Ø¶Ø±ÙØ¨Ø© + nLPD + GDPRØ Ù
ÙØ²Ø§ÙÙØ© 20,000 ÙØ±ÙÙ â Ù
عتÙ
دة بÙÙØ±Ø© ÙØ§ØØ¯Ø©.",s3t:"ÙØ¸ÙÙ Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",s3d:"اختر ÙØ±ÙÙÙ Ù
٠اÙÙ
تجر: اÙÙ
Ø¨ÙØ¹Ø§ØªØ Ø§ÙØ¯Ø¹Ù
Ø Ø§ÙØªØ³ÙÙÙØ اÙÙ
اÙÙØ©Ø Ø§ÙØ¨ØØ«Ø اÙÙØ§ÙÙÙ ÙØºÙØ±ÙØ§. ÙÙ ÙÙÙÙ Ø¬Ø§ÙØ² ÙÙØ±ÙØ§Ø 24/7Ø Ø¯ÙÙ ØªØ¯Ø±ÙØ¨.",s3e:"Ù
ثاÙ: ÙØ±ÙÙ Ù
Ù 5 ÙÙÙØ§Ø¡ ÙØ¹ÙÙØ¶ عبء عÙ
Ù 8 Ù
ÙØ¸ÙÙÙ Ù
بتدئÙÙ Ø¨Ø¯ÙØ§Ù
ÙØ§Ù
Ù.",s4t:"اربط Ø£Ø¯ÙØ§ØªÙ",s4d:"93 Ù
ÙØµÙÙØ§ Ù
تاØÙا (CRMØ Ø§ÙØ¨Ø±ÙØ¯Ø Ø§ÙÙ
ØØ§Ø³Ø¨Ø©Ø Ø§ÙØ¯Ø¹Ù
Ø Ø§ÙØ³ØØ§Ø¨Ø©Ø Ø§ÙØªØ¬Ø§Ø±Ø© Ø§ÙØ¥ÙÙØªØ±ÙÙÙØ©). ÙØ¹Ù
Ù Ù
ÙØ¸ÙÙÙ Ø§ÙØ°ÙÙÙÙ Ù
باشرة داخ٠Ù
ÙØ¸ÙÙ
ØªÙ Ø§ÙØØ§ÙÙØ©.",s4e:"Ù
ثاÙ: ربط Stripe + HubSpot + Gmail Ù٠أÙÙ Ù
Ù 10 Ø¯ÙØ§Ø¦ÙØ Ø¨ÙØ§Ùات Ø§ÙØ§Ø¹ØªÙ
اد Ù
Ø´ÙÙØ±Ø© AES-256.",s5t:"Ø§ÙØ´Ø± Ø§ÙØ´Ø±ÙØ©",s5d:"ÙØªÙ
تجÙÙØ² Ø§ÙØ¹ÙاÙ
Ø© ÙØ§ÙÙØ·Ø§Ù ÙØ§ÙÙ
ÙÙØ¹ ÙØ§ÙبÙÙØ© Ø§ÙØ³ØØ§Ø¨ÙØ© ÙØ§ÙÙ
دÙÙØ¹Ø§Øª ÙØ§ÙعÙ
ÙÙØ§Øª تÙÙØ§Ø¦ÙÙØ§. تبÙÙ Ù٠عÙ
ÙÙØ© ØºÙØ± ÙØ§Ø¨ÙØ© ÙÙØªØ±Ø§Ø¬Ø¹ خاضعة ÙÙ
ÙØ§ÙÙØ© Ø¨Ø´Ø±ÙØ©.",s5e:"Ù
ثاÙ: اÙÙØ·Ø§Ù + اÙÙ
ÙÙØ¹ + تدÙÙ Ø§ÙØ¯Ùع Ø®ÙØ§Ù 24 ساعة بدÙÙØ§ Ù
Ù 6 Ø£Ø³Ø§Ø¨ÙØ¹ Ù
ع ÙÙØ§ÙØ©.",s6t:"ØªÙØ³Ùع عاÙÙ
ÙÙØ§",s6d:"Ù
تعدد Ø§ÙØ¯ÙÙ ÙØ§ÙعÙ
ÙØ§Øª ÙØ§ÙÙØºØ§Øª. ÙØªÙÙØ¹ اÙÙÙÙÙ Ø§ÙØ±Ø¦ÙØ³Ù Ø§ÙØ³ÙÙÙØ© ÙØ§ÙÙ
خاطر ÙØ§ÙÙ
Ø¨ÙØ¹Ø§Øª ÙØ§ØØªÙاجات Ø§ÙØªÙظÙÙ Ø«Ù
ÙÙÙÙØ¶ Ø§ÙØ¹Ù
٠إÙÙ ÙÙÙØ§Ø¦Ù.",s6e:"Ù
ثاÙ: ÙØªØ سÙ٠ثاÙÙ (Ø£ÙÙ
اÙÙØ§) Ù
ع Ù
Ø¹Ø§ÙØ¬Ø© اÙÙØ§ÙÙÙ ÙØ§Ùضرائب ÙØ§ÙØªÙØ·Ù٠تÙÙØ§Ø¦ÙÙØ§.",saveT:"Ø§ÙØªÙÙÙØ© â ÙÙ
ا تÙÙØ±Ù",saveS:"Ù
ÙØ§Ø±ÙØ© بÙÙ ÙØ±Ù٠بشر٠تÙÙÙØ¯Ù ÙÙÙØ³ اÙÙØ¸Ø§Ø¦Ù Ø¨ÙØ§Ø³Ø·Ø© Ù
ÙØ¸ÙÙ Infernex Ø§ÙØ°ÙÙÙÙ. Ø£Ø±ÙØ§Ù
Ø´ÙØ±ÙØ© ÙÙÙ ÙØ¸ÙÙØ©.",colRole:"اÙÙØ¸ÙÙØ©",colHuman:"ÙØ±Ù٠بشر٠/ Ø´ÙØ±",colIx:"Infernex / Ø´ÙØ±",colSave:"Ø§ÙØªÙÙÙØ±",r1:"اÙÙ
Ø¨ÙØ¹Ø§Øª ÙØ§ÙØ§Ø³ØªÙØ·Ø§Ø¨",r2:"دعÙ
Ø§ÙØ¹Ù
ÙØ§Ø¡ 24/7",r3:"Ø§ÙØªØ³ÙÙÙ ÙØ§ÙÙ
ØØªÙÙ",r4:"اÙÙ
اÙÙØ© ÙØ§ÙÙÙØªØ±Ø©",r5:"Ø§ÙØ¨ØØ« ÙØ§ÙعÙ
ÙÙØ§Øª",totalRow:"Ø§ÙØ¥Ø¬Ù
اÙÙ",note:"ØªØ³ØªÙØ¯ Ø§ÙØªÙاÙÙÙ Ø§ÙØ¨Ø´Ø±ÙØ© Ø¥ÙÙ Ù
ØªÙØ³Ø· Ø§ÙØ±Ùاتب Ø§ÙØ£ÙØ±ÙØ¨ÙØ© Ø§ÙØ¥Ø¬Ù
اÙÙØ© (Ù
بتدئ Ø¥ÙÙ Ù
ØªÙØ³Ø·
811). تشÙ
Ù ØªÙØ§ÙÙÙ Infernex Ø§ÙØ§Ø´ØªØ±Ø§Ù ÙØ§ÙØ§Ø³ØªØ¯ÙØ§Ù اÙÙ
ÙØ¬Ù٠إÙ٠أرخص Ù
ØØ±Ù ÙÙØ¨Ù Ø§ØªÙØ§ÙÙØ© Ø§ÙØ®Ø¯Ù
Ø© â Ø£Ø±Ø®Øµ Ø¨ÙØ³Ø¨Ø© تص٠إÙÙ 95% Ù
Ù Ø§ÙØ§Ø³ØªØ¯Ø¹Ø§Ø¡ اÙÙ
باشر ÙÙÙÙ
اذج اÙÙ
ØªÙØ¯Ù
Ø©.",ctaT:"Ø¬Ø§ÙØ² ÙØ¥Ø·ÙØ§Ù Ø´Ø±ÙØªÙØ",ctaS:"Ø£ÙØ´Ø¦ Ø£ÙÙ Ù
ÙØ¸Ù Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù Ù
جاÙÙØ§Ø Ø«Ù
Ø§ÙØ´Ø± Ø´Ø±ÙØªÙ بÙÙØ±Ø© ÙØ§ØØ¯Ø©.",ctaB:"Ø¥ÙØ´Ø§Ø¡ Ø´Ø±ÙØªÙ"}}},Dr={en:{wfp:{f:{employees:"AI employees",languages:"Languages available",website:"Company website included",crm:"Built-in CRM",connectors:"Connectors & integrations",automations:"Automations & workflows",orchestrator:"Master AI Orchestrator",tokens:"Tokens included / month",seats:"Human seats",api:"Inference API (OpenAI-compatible)",analytics:"ROI analytics dashboard",security:"Security & compliance",support:"Support"},v:{unlimited:"Unlimited",custom:"Custom volume",yes:"Included",no:"Not included",limited:"Limited access",basic:"Basic",advanced:"Advanced",full:"Full access",lang1:"1 language (auto-detected from client IP)",lang3:"3 languages of your choice",lang9:"All Infernex languages (9)",web1:"1-page site + basic SEO",web15:"Up to 15 pages + SEO & hreflang",webUnlimited:"Unlimited pages + custom domains",conn5:"Email + Slack (5 connectors)",conn88:"Salesforce, HubSpot, Shopify (88+)",connAll:"All connectors + private custom ones",tok500k:"500K tokens",tok5m:"5M tokens",yesPrivate:"Included + private endpoints",sec1:"RLS + TLS 1.3",sec2:"AI Safety Layer + approvals",sec3:"SSO/SAML, VPC/on-prem, audit logs",community:"Community",priority:"Priority (24h)",dedicated:"Dedicated manager + 99.99% SLA"},tiers:{starter:{name:"Starter",desc:"For small teams launching their first AI workforce."},business:{name:"Business",desc:"For growing companies scaling AI operations."},enterprise:{name:"Enterprise",desc:"For organizations with mission-critical deployments."}},compareTitle:"Full plan comparison",compareSubtitle:"Everything included in each plan, side by side.",colFeature:"Feature",upgradeTitle:"Upgrade whenever you grow",upgradeText:"Start on Starter, move to Business or Enterprise in one click. Your AI employees, data, CRM and website are kept â only limits and languages expand. Prorated billing, no migration, cancel anytime.",upgradeCta:"See upgrade path",popular:"Most popular",perMonth:"/month",customPrice:"Custom",ctaStart:"Start Free Trial",ctaContact:"Contact Sales",ctaDetail:"See full plan details",excl:{title:"External costs not included in your plan",intro:"Full transparency: your Infernex plan covers the platform (AI employees, CRM, website, automations, inference API). The following third-party costs stay on your side.",items:["Domain name and its yearly renewal (~$10â50/year) â Infernex does not resell domains.","External hosting or professional email if you do not use the one included.","Paid third-party connectors: the subscriptions of the tools you plug in (HubSpot, Salesforce, paid Slack, Shopify, LinkedIn Sales Navigatorâ¦). We provide the connection, not the licence.","Third-party API keys you bring (BYOK) and their own consumption.","Token overage beyond your plan quota, billed on usage.","Local taxes/VAT, currency conversion and bank fees.","Optional services: assisted migration, custom connector development, reinforced SLA.","Contracts, licences and legal obligations required for your company to be registered, compliant and operational in your country (company registration, insurance, local taxes, GDPR compliance, professional licences, etc.) are not included and remain at your expense."],note:"Your company's final budget depends on your number of iterations, active AI employees and token consumption. The plan price covers the platform, not your external spend.",disclaimer:"The user is solely responsible for the actions, decisions and content generated through the Infernex AI platform. Infernex provides 24/7 technical support by email but disclaims any liability for misuse of the service, any use that does not comply with applicable laws or our terms of use, and for the commercial, financial or legal consequences that may result."},home:{badge:"PLANS & PRICING",title:"Simple pricing, from $99/month",subtitle:"Know exactly what you pay before you build. Starter, Business or Enterprise â upgrade at any time.",cta:"Compare all plans"}}},fr:{wfp:{f:{employees:"Employés IA",languages:"Langues disponibles",website:"Site web d'entreprise inclus",crm:"CRM intégré",connectors:"Connecteurs & intégrations",automations:"Automatisations & workflows",orchestrator:"Agent IA Maître (Orchestrateur)",tokens:"Jetons inclus / mois",seats:"Utilisateurs humains",api:"API d'inférence (compatible OpenAI)",analytics:"Tableau de bord ROI",security:"Sécurité & conformité",support:"Support"},v:{unlimited:"Illimité",custom:"Volume sur mesure",yes:"Inclus",no:"Non inclus",limited:"Accès limité",basic:"Basique",advanced:"Avancé",full:"Accès complet",lang1:"1 langue (détectée selon l'IP du client)",lang3:"3 langues au choix",lang9:"Toutes les langues Infernex (9)",web1:"Site 1 page + SEO de base",web15:"Jusqu'à 15 pages + SEO & hreflang",webUnlimited:"Pages illimitées + domaines personnalisés",conn5:"Email + Slack (5 connecteurs)",conn88:"Salesforce, HubSpot, Shopify (88+)",connAll:"Tous les connecteurs + connecteurs privés",tok500k:"500K
811jetons",tok5m:"5M jetons",yesPrivate:"Inclus + endpoints privés",sec1:"RLS + TLS 1.3",sec2:"AI Safety Layer + validations",sec3:"SSO/SAML, VPC/on-prem, journaux d'audit",community:"Communauté",priority:"Prioritaire (24h)",dedicated:"Responsable dédié + SLA 99,99 %"},tiers:{starter:{name:"Starter",desc:"Pour les petites équipes qui lancent leur première workforce IA."},business:{name:"Business",desc:"Pour les entreprises en croissance qui industrialisent l'IA."},enterprise:{name:"Entreprise",desc:"Pour les organisations avec des déploiements critiques."}},compareTitle:"Comparatif complet des plans",compareSubtitle:"Tout ce qui est inclus dans chaque plan, côte à côte.",colFeature:"Fonctionnalité",upgradeTitle:"Ãvoluez quand vous grandissez",upgradeText:"Commencez en Starter, passez en Business ou Entreprise en un clic. Vos employés IA, données, CRM et site web sont conservés â seules les limites et les langues augmentent. Facturation au prorata, aucune migration, annulation à tout moment.",upgradeCta:"Voir le parcours d'upgrade",popular:"Le plus choisi",perMonth:"/mois",customPrice:"Sur devis",ctaStart:"Démarrer l'essai gratuit",ctaContact:"Contacter les ventes",ctaDetail:"Voir le détail du plan",excl:{title:"Frais externes non inclus dans votre plan",intro:"Transparence totale : votre plan Infernex couvre la plateforme (employés IA, CRM, site web, automatisations, API d'inférence). Les frais tiers ci-dessous restent à votre charge.",items:["Nom de domaine et son renouvellement annuel (~10â50 $/an) â Infernex ne revend pas de domaines.","Hébergement ou e-mail professionnel externe si vous n'utilisez pas celui inclus.","Connecteurs tiers payants : les abonnements des outils que vous branchez (HubSpot, Salesforce, Slack payant, Shopify, LinkedIn Sales Navigatorâ¦). Nous fournissons la connexion, pas la licence.","Clés API tierces que vous apportez (BYOK) et leur propre consommation.","Dépassement de jetons au-delà du quota du plan, facturé à l'usage.","Taxes/TVA locales, frais de change et frais bancaires.","Prestations optionnelles : migration assistée, connecteur sur mesure, SLA renforcé.","Les contrats, licences et obligations légales obligatoires pour que votre entreprise soit immatriculée, conforme et opérationnelle dans votre pays (registre du commerce, assurances, fiscalité locale, RGPD, licences professionnelles, etc.) ne sont pas inclus dans le plan et restent à votre charge."],note:"Le budget final de votre entreprise dépend du nombre d'itérations, d'employés IA actifs et de votre consommation de jetons. Le prix du plan couvre la plateforme, pas vos dépenses externes.",disclaimer:"L'utilisateur est seul responsable des actions, décisions et contenus générés via la plateforme Infernex AI. Infernex fournit une assistance technique 24h/24 et 7j/7 par e-mail, mais décline toute responsabilité en cas de mauvaise utilisation du service, d'usage non conforme aux lois applicables ou à nos conditions d'utilisation, ainsi que des conséquences commerciales, financières ou légales qui en découleraient."},home:{badge:"PLANS & TARIFS",title:"Tarifs simples, dès 99 $/mois",subtitle:"Sachez exactement ce que vous payez avant de construire. Starter, Business ou Entreprise â évoluez à tout moment.",cta:"Comparer tous les plans"}}},es:{wfp:{f:{employees:"Empleados IA",languages:"Idiomas disponibles",website:"Sitio web de empresa incluido",crm:"CRM integrado",connectors:"Conectores e integraciones",automations:"Automatizaciones y flujos",orchestrator:"Agente IA Maestro (Orquestador)",tokens:"Tokens incluidos / mes",seats:"Usuarios humanos",api:"API de inferencia (compatible OpenAI)",analytics:"Panel de ROI",security:"Seguridad y cumplimiento",support:"Soporte"},v:{unlimited:"Ilimitado",custom:"Volumen a medida",yes:"Incluido",no:"No incluido",limited:"Acceso limitado",basic:"Básico",advanced:"Avanzado",full:"Acceso completo",lang1:"1 idioma (detectado por la IP del cliente)",lang3:"3 idiomas a elegir",lang9:"Todos los idiomas de Infernex (9)",web1:"Sitio de 1 página + SEO básico",web15:"Hasta 15 páginas + SEO y hreflang",webUnlimited:"Páginas
811ilimitadas + dominios propios",conn5:"Email + Slack (5 conectores)",conn88:"Salesforce, HubSpot, Shopify (88+)",connAll:"Todos los conectores + privados a medida",tok500k:"500K tokens",tok5m:"5M tokens",yesPrivate:"Incluido + endpoints privados",sec1:"RLS + TLS 1.3",sec2:"AI Safety Layer + aprobaciones",sec3:"SSO/SAML, VPC/on-prem, registros de auditorÃa",community:"Comunidad",priority:"Prioritario (24h)",dedicated:"Gestor dedicado + SLA 99,99 %"},tiers:{starter:{name:"Starter",desc:"Para equipos pequeños que lanzan su primera plantilla IA."},business:{name:"Business",desc:"Para empresas en crecimiento que escalan operaciones IA."},enterprise:{name:"Enterprise",desc:"Para organizaciones con despliegues crÃticos."}},compareTitle:"Comparativa completa de planes",compareSubtitle:"Todo lo incluido en cada plan, lado a lado.",colFeature:"Función",upgradeTitle:"Sube de plan cuando crezcas",upgradeText:"Empieza en Starter y pasa a Business o Enterprise en un clic. Tus empleados IA, datos, CRM y sitio web se conservan: solo aumentan los lÃmites y los idiomas. Facturación prorrateada, sin migración, cancela cuando quieras.",upgradeCta:"Ver ruta de mejora",popular:"Más elegido",perMonth:"/mes",customPrice:"A medida",ctaStart:"Iniciar prueba gratuita",ctaContact:"Contactar ventas",ctaDetail:"Ver el detalle del plan",excl:{title:"Costes externos no incluidos en tu plan",intro:"Transparencia total: tu plan Infernex cubre la plataforma (empleados IA, CRM, sitio web, automatizaciones, API de inferencia). Los siguientes costes de terceros corren por tu cuenta.",items:["Nombre de dominio y su renovación anual (~10â50 $/año) â Infernex no revende dominios.","Alojamiento o correo profesional externo si no usas el incluido.","Conectores de terceros de pago: las suscripciones de las herramientas que conectes (HubSpot, Salesforce, Slack de pago, Shopify, LinkedIn Sales Navigatorâ¦). Aportamos la conexión, no la licencia.","Claves API de terceros que aportes (BYOK) y su propio consumo.","Exceso de tokens por encima de la cuota del plan, facturado por uso.","Impuestos/IVA locales, cambio de divisa y comisiones bancarias.","Servicios opcionales: migración asistida, conector a medida, SLA reforzado.","Los contratos, licencias y obligaciones legales necesarios para que su empresa esté registrada, en conformidad y operativa en su paÃs (registro mercantil, seguros, fiscalidad local, RGPD, licencias profesionales, etc.) no están incluidos y corren por su cuenta."],note:"El presupuesto final de tu empresa depende del número de iteraciones, empleados IA activos y consumo de tokens. El precio del plan cubre la plataforma, no tus gastos externos.",disclaimer:"El usuario es el único responsable de las acciones, decisiones y contenidos generados a través de la plataforma Infernex AI. Infernex ofrece soporte técnico 24/7 por correo electrónico, pero declina toda responsabilidad por un uso indebido del servicio, por un uso no conforme con las leyes aplicables o con nuestras condiciones de uso, asà como por las consecuencias comerciales, financieras o legales que puedan derivarse."},home:{badge:"PLANES Y PRECIOS",title:"Precios simples, desde 99 $/mes",subtitle:"Sepa exactamente lo que paga antes de construir. Starter, Business o Enterprise: mejore cuando quiera.",cta:"Comparar todos los planes"}}},de:{wfp:{f:{employees:"KI-Mitarbeiter",languages:"Verfügbare Sprachen",website:"Firmen-Website inklusive",crm:"Integriertes CRM",connectors:"Konnektoren & Integrationen",automations:"Automatisierungen & Workflows",orchestrator:"Master-KI-Orchestrator",tokens:"Inkludierte Tokens / Monat",seats:"Nutzerplätze",api:"Inferenz-API (OpenAI-kompatibel)",analytics:"ROI-Dashboard",security:"Sicherheit & Compliance",support:"Support"},v:{unlimited:"Unbegrenzt",custom:"Individuelles Volumen",yes:"Inklusive",no:"Nicht enthalten",limited:"Eingeschränkter Zugriff",basic:"Basis",advanced:"Erweitert",full:"Voller Zugriff",lang1:"1 Sprache (automatisch per Kunden-IP)",lang3:"3 Sprachen nach Wahl",lang9:"Alle Infernex-Sprachen (9)",web1:"1-Seiten-Website + Basis-SEO",web15:"Bis zu 15 Seiten + SEO & hreflang",webUnlimited:"Unbegrenzte Seiten + eigene Domains",conn5:"E-Mail + Slack (5 Konnektoren)",conn88:"Salesforce, HubSpot, Shopify (88+)",connAll:"Alle Konnektoren + private Sonderanbindungen",tok500k:"500K Tokens",tok5m:"5M Tokens",yesPrivate:"Inklusive + private Endpunkte",sec1:"RLS + TLS 1.3",sec2:"AI Safety Layer + Freigaben",sec3:"SSO/SAML, VPC/on-prem, Audit-Logs",community:"Community",priority:"Priorität (24 Std.)",dedicated:"Dedizierter Manager + 99,99 % SLA"},tiers:{starter:{name:"Starter",desc:"Für kleine Teams mit ihrer ersten KI-Belegschaft."},business:{name:"Business",desc:"Für wachsende Unternehmen, die KI skalieren."},enterprise:{name:"Enterprise",desc:"Für Organisationen mit unternehmenskritischen Deployments."}},compareTitle:"Vollständiger Plan-Vergleich",compareSubtitle:"Alle Leistungen jedes Plans im direkten Vergleich.",colFeature:"Funktion",upgradeTitle:"Upgrade, wann immer Sie wac
811hsen",upgradeText:"Starten Sie mit Starter und wechseln Sie per Klick zu Business oder Enterprise. KI-Mitarbeiter, Daten, CRM und Website bleiben erhalten â nur Limits und Sprachen wachsen. Anteilige Abrechnung, keine Migration, jederzeit kündbar.",upgradeCta:"Upgrade-Pfad ansehen",popular:"Am beliebtesten",perMonth:"/Monat",customPrice:"Auf Anfrage",ctaStart:"Kostenlos testen",ctaContact:"Vertrieb kontaktieren",ctaDetail:"Tarifdetails ansehen",excl:{title:"Externe Kosten, die nicht im Tarif enthalten sind",intro:"Volle Transparenz: Ihr Infernex-Tarif deckt die Plattform ab (KI-Mitarbeiter, CRM, Website, Automatisierungen, Inferenz-API). Die folgenden Drittkosten tragen Sie selbst.",items:["Domainname und jährliche Verlängerung (~10â50 $/Jahr) â Infernex verkauft keine Domains.","Externes Hosting oder professionelle E-Mail, falls Sie die enthaltene nicht nutzen.","Kostenpflichtige Drittanbieter-Konnektoren: die Abos der Tools, die Sie anbinden (HubSpot, Salesforce, Slack Paid, Shopify, LinkedIn Sales Navigatorâ¦). Wir liefern die Verbindung, nicht die Lizenz.","Eigene Drittanbieter-API-Schlüssel (BYOK) und deren Verbrauch.","Token-Ãberschreitung über das Tarifkontingent hinaus, nutzungsbasiert abgerechnet.","Lokale Steuern/MwSt., Währungsumrechnung und Bankgebühren.","Optionale Leistungen: begleitete Migration, individueller Konnektor, verstärktes SLA.","Verträge, Lizenzen und rechtliche Pflichten, die erforderlich sind, damit Ihr Unternehmen in Ihrem Land eingetragen, konform und betriebsbereit ist (Handelsregister, Versicherungen, lokale Steuern, DSGVO, Berufslizenzen usw.), sind nicht enthalten und gehen zu Ihren Lasten."],note:"Das Endbudget Ihres Unternehmens hängt von der Anzahl Ihrer Iterationen, aktiven KI-Mitarbeiter und Ihrem Token-Verbrauch ab. Der Tarifpreis deckt die Plattform, nicht Ihre externen Ausgaben.",disclaimer:"Der Nutzer ist allein verantwortlich für die über die Infernex-AI-Plattform erzeugten Handlungen, Entscheidungen und Inhalte. Infernex bietet technischen Support rund um die Uhr per E-Mail, lehnt jedoch jede Haftung für missbräuchliche Nutzung des Dienstes, für eine nicht gesetzeskonforme oder gegen unsere Nutzungsbedingungen verstoÃende Nutzung sowie für daraus resultierende geschäftliche, finanzielle oder rechtliche Folgen ab."},home:{badge:"PLÃNE & PREISE",title:"Einfache Preise, ab 99 $/Monat",subtitle:"Wissen Sie genau, was Sie zahlen, bevor Sie bauen. Starter, Business oder Enterprise â jederzeit upgraden.",cta:"Alle Pläne vergleichen"}}},pt:{wfp:{f:{employees:"Funcionários IA",languages:"Idiomas disponÃveis",website:"Site da empresa incluÃdo",crm:"CRM integrado",connectors:"Conectores e integrações",automations:"Automações e fluxos",orchestrator:"Agente IA Mestre (Orquestrador)",tokens:"Tokens incluÃdos / mês",seats:"Utilizadores humanos",api:"API de inferência (compatÃvel OpenAI)",analytics:"Painel de ROI",security:"Segurança e conformidade",support:"Suporte"},v:{unlimited:"Ilimitado",custom:"Volume personalizado",yes:"IncluÃdo",no:"Não incluÃdo",limited:"Acesso limitado",basic:"Básico",advanced:"Avançado",full:"Acesso completo",lang1:"1 idioma (detetado pelo IP do cliente)",lang3:"3 idiomas à escolha",lang9:"Todos os idiomas Infernex (9)",web1:"Site de 1 página + SEO básico",web15:"Até 15 páginas + SEO e hreflang",webUnlimited:"Páginas ilimitadas + domÃnios próprios",conn5:"Email + Slack (5 conectores)",conn88:"Salesforce, HubSpot, Shopify (88+)",connAll:"Todos os conectores + privados à medida",tok500k:"500K tokens",tok5m:"5M tokens",yesPrivate:"IncluÃdo + endpoints privados",sec1:"RLS + TLS 1.3",sec2:"AI Safety Layer + aprovações",sec3:"SSO/SAML, VPC/on-prem, registos de auditoria",community:"Comunidade",priority:"Prioritário (24h)",dedicated:"Gestor dedicado + SLA 99,99 %"},tiers:{starter:{name:"Starter",desc:"Para equipas pequenas que lançam a sua primeira força de trabalho IA."},business:{name:"Business",desc:"Para empresas em crescimento que escalam operações de IA."},enterprise:{name:"Enterprise",desc:"Para organizações com implementações crÃticas."}},compareTitle:"Comparativo completo dos planos",compareSubtitle:"Tudo o que está incluÃdo em c
811ada plano, lado a lado.",colFeature:"Funcionalidade",upgradeTitle:"Evolua quando crescer",upgradeText:"Comece no Starter e passe para Business ou Enterprise num clique. Os seus funcionários IA, dados, CRM e site são mantidos â só os limites e idiomas aumentam. Faturação proporcional, sem migração, cancele quando quiser.",upgradeCta:"Ver percurso de upgrade",popular:"Mais escolhido",perMonth:"/mês",customPrice:"Sob consulta",ctaStart:"Começar teste grátis",ctaContact:"Falar com vendas",ctaDetail:"Ver o detalhe do plano",excl:{title:"Custos externos não incluÃdos no seu plano",intro:"Transparência total: o seu plano Infernex cobre a plataforma (funcionários IA, CRM, site, automações, API de inferência). Os custos de terceiros abaixo ficam a seu cargo.",items:["Nome de domÃnio e a renovação anual (~10â50 $/ano) â a Infernex não revende domÃnios.","Alojamento ou e-mail profissional externo, se não usar o incluÃdo.","Conectores de terceiros pagos: as subscrições das ferramentas que ligar (HubSpot, Salesforce, Slack pago, Shopify, LinkedIn Sales Navigatorâ¦). Fornecemos a ligação, não a licença.","Chaves de API de terceiros que traga (BYOK) e o respetivo consumo.","Excedente de tokens acima da quota do plano, faturado ao uso.","Impostos/IVA locais, câmbio e taxas bancárias.","Serviços opcionais: migração assistida, conector à medida, SLA reforçado.","Os contratos, licenças e obrigações legais necessários para que a sua empresa esteja registada, em conformidade e operacional no seu paÃs (registo comercial, seguros, fiscalidade local, RGPD, licenças profissionais, etc.) não estão incluÃdos e ficam a seu cargo."],note:"O orçamento final da sua empresa depende do número de iterações, funcionários IA ativos e consumo de tokens. O preço do plano cobre a plataforma, não as suas despesas externas.",disclaimer:"O utilizador é o único responsável pelas ações, decisões e conteúdos gerados através da plataforma Infernex AI. A Infernex disponibiliza suporte técnico 24/7 por e-mail, mas declina qualquer responsabilidade por utilização indevida do serviço, por qualquer utilização não conforme com as leis aplicáveis ou com os nossos termos de utilização, bem como pelas consequências comerciais, financeiras ou legais daà resultantes."},home:{badge:"PLANOS E PREÃOS",title:"Preços simples, a partir de 99 $/mês",subtitle:"Saiba exatamente quanto paga antes de construir. Starter, Business ou Enterprise â evolua quando quiser.",cta:"Comparar todos os planos"}}},zh:{wfp:{f:{employees:"AI åå·¥æ°é",languages:"å¯ç¨è¯è¨",website:"å
å«ä¼ä¸ç½ç«",crm:"å
ç½® CRM",connectors:"è¿æ¥å¨ä¸éæ",automations:"èªå¨åä¸å·¥ä½æµ",orchestrator:"ä¸»æ§ AI ç¼æå¨",tokens:"æ¯æå
å« Token",seats:"人类å¸ä½",api:"æ¨ç APIï¼å
¼å®¹ OpenAIï¼",analytics:"ROI åæé¢æ¿",security:"å®å
¨ä¸åè§",support:"æ¯ææå¡"},v:{unlimited:"æ éå¶",custom:"å®å¶é¢åº¦",yes:"å
å«",no:"ä¸å
å«",limited:"æé访é®",basic:"åºç¡ç",advanced:"è¿é¶ç",full:"宿´è®¿é®",lang1:"1 ç§è¯è¨ï¼æå®¢æ· IP èªå¨è¯å«ï¼",lang3:"ä»»é 3 ç§è¯è¨",lang9:"Infernex å
¨é¨è¯è¨ï¼9 ç§ï¼",web1:"å页ç½ç« + åºç¡ SEO",web15:"æå¤ 15 页 + SEO ä¸ hreflang",webUnlimited:"页颿 é + èªå®ä¹åå",conn5:"Email + Slackï¼5 ä¸ªè¿æ¥å¨ï¼",conn88:"SalesforceãHubSpotãShopifyï¼88+ï¼",connAll:"å
¨é¨è¿æ¥å¨ + ç§æå®å¶",tok500k:"50 ä¸ Token",tok5m:"500 ä¸ Token",yesPrivate:"å
å« + ç§æç«¯ç¹",sec1:"RLS + TLS 1.3",sec2:"AI å®å
¨å± + å®¡æ¹æµç¨",sec3:"SSO/SAMLãVPC/æ¬å°é¨ç½²ã审计æ¥å¿",community:"ç¤¾åºæ¯æ",priority:"ä¼å
æ¯æï¼24 å°æ¶ï¼",dedicated:"ä¸å±ç»ç + 99.99% SLA"},tiers:{starter:{name:"Starter",desc:"éå馿¬¡é¨ç½² AI å¢éçå°åå¢éã"},business:{name:"Business",desc:"éåæ©å± AI è¿è¥çæé¿åå
¬å¸ã"},enterprise:{name:"Enterprise",desc:"éåå
³é®ä¸å¡é¨ç½²ç大åç»ç»ã"}},compareTitle:"宿´æ¹æ¡å¯¹æ¯",compareSubtitle:"éé¡¹å¯¹æ¯æ¯ä¸ªæ¹æ¡å
å«çå
容ã",colFeature:"åè½",upgradeTitle:"éä¸å¡å¢é¿éæ¶å级",upgradeText:"ä» Starter å¼å§ï¼ä¸é®åçº§å° Business æ EnterpriseãAI åå·¥ãæ°æ®ãCRM åç½ç«å
¨é¨ä¿çï¼ä»
é¢åº¦ä¸è¯è¨æ©å±ãææ¯ä¾è®¡è´¹ï¼æ éè¿ç§»ï¼å¯éæ¶åæ¶ã",upgradeCta:"æ¥çå级路å¾",popular:"æå欢è¿",perMonth:"/æ",customPrice:"å®å¶æ¥ä»·",ctaStart:"å¼å§å
è´¹è¯ç¨",ctaContact:"èç³»éå®",ctaDetail:"æ¥çå¥é¤è¯¦æ
",excl:{title:"å¥é¤æªå
å«çå¤é¨è´¹ç¨",intro:"å®å
¨éæï¼Infernex å¥é¤æ¶µç平尿¬èº«ï¼AI åå·¥ãCRMãç½ç«ãèªå¨åãæ¨ç APIï¼ã以ä¸ç¬¬ä¸æ¹è´¹ç¨ç±æ¨æ¿æ
ã",items:["åååå
¶å¹´åº¦ç»è´¹ï¼çº¦ 10â50 ç¾å
/å¹´ï¼ââInfernex ä¸è½¬å®ååã","å¦ä¸ä½¿ç¨å
ç½®æå¡ï¼å¤é¨æç®¡æä¼ä¸é®ç®±è´¹ç¨ã","ä»è´¹ç¬¬ä¸æ¹è¿æ¥å¨ï¼æ¨æ¥å
¥å·¥å
·ç订é
è´¹ï¼HubSpotãSalesforceãä»è´¹ SlackãShopifyãLinkedIn Sales Navigator çï¼ãæä»¬æä¾è¿æ¥ï¼ä¸æä¾è®¸å¯è¯ã","æ¨èªå¸¦çç¬¬ä¸æ¹ API å¯é¥ï¼BYOKï¼åå
¶æ¶èã","è¶
åºå¥é¤é¢åº¦ç Token ç¨éï¼æç¨é计费ã","å½å°ç¨è´¹/å¢å¼ç¨ãæ±å
åé¶è¡æç»è´¹ã","å¯éæå¡ï¼åå©è¿ç§»ãå
811®å¶è¿æ¥å¨ãå¢å¼º SLAã","使æ¨çå
¬å¸å¨æå¨å½å®¶å®ææ³¨åãåè§å¹¶æ£å¸¸è¿è¥æéçååã许å¯è¯åæ³å¾ä¹å¡ï¼å·¥å注åãä¿é©ãå½å°ç¨å¡ãGDPR åè§ãèä¸è®¸å¯çï¼ä¸å
å«å¨å¥é¤å
ï¼éç±æ¨èªè¡æ¿æ
ã"],note:"æ¨å
¬å¸çæç»é¢ç®åå³äºè¿ä»£æ¬¡æ°ãæ´»è· AI åå·¥æ°éå Token æ¶èãå¥é¤ä»·æ ¼æ¶µç平尿¬èº«ï¼ä¸å«æ¨çå¤é¨æ¯åºã",disclaimer:"ç¨æ·å¯¹éè¿ Infernex AI å¹³å°äº§ççæææä½ãå³çåå
å®¹æ¿æ
å
¨é¨è´£ä»»ãInfernex æä¾ 7Ã24 å°æ¶çµåé®ä»¶ææ¯æ¯æï¼ä½å¯¹æå¡çä¸å½ä½¿ç¨ãä¸ç¬¦åéç¨æ³å¾ææä»¬ä½¿ç¨æ¡æ¬¾ç使ç¨è¡ä¸ºï¼ä»¥åç±æ¤äº§ççåä¸ãè´¢å¡ææ³å¾åææ¦ä¸è´è´£ã"},home:{badge:"æ¹æ¡ä¸ä»·æ ¼",title:"ç®åå®ä»·ï¼æ¯æ 99 ç¾å
èµ·",subtitle:"å¨å¼å§æå»ºåå°±æ¸
æ¥ç¥éè´¹ç¨ãStarterãBusiness æ Enterpriseï¼å¯éæ¶å级ã",cta:"对æ¯å
¨é¨æ¹æ¡"}}},ja:{wfp:{f:{employees:"AI 徿¥å¡æ°",languages:"å©ç¨å¯è½ãªè¨èª",website:"伿¥ã¦ã§ããµã¤ãè¾¼ã¿",crm:"å
èµ CRM",connectors:"ã³ãã¯ã¿ã¨é£æº",automations:"èªååã¨ã¯ã¼ã¯ããã¼",orchestrator:"ãã¹ã¿ã¼ AI ãªã¼ã±ã¹ãã¬ã¼ã¿ã¼",tokens:"æéãã¼ã¯ã³",seats:"人éã®ã·ã¼ãæ°",api:"æ¨è« APIï¼OpenAI äºæï¼",analytics:"ROI ããã·ã¥ãã¼ã",security:"ã»ãã¥ãªãã£ã¨ã³ã³ãã©ã¤ã¢ã³ã¹",support:"ãµãã¼ã"},v:{unlimited:"ç¡å¶é",custom:"ã«ã¹ã¿ã é",yes:"å«ã",no:"å«ã¾ãªã",limited:"å¶éä»ãã¢ã¯ã»ã¹",basic:"ãã¼ã·ãã¯",advanced:"ã¢ããã³ã¹",full:"ãã«ã¢ã¯ã»ã¹",lang1:"1 è¨èªï¼é¡§å®¢ IP ããèªåå¤å®ï¼",lang3:"ä»»æã® 3 è¨èª",lang9:"Infernex ã®å
¨è¨èªï¼9 è¨èªï¼",web1:"1 ãã¼ã¸ãµã¤ã + åºæ¬ SEO",web15:"æå¤§ 15 ãã¼ã¸ + SEOã»hreflang",webUnlimited:"ç¡å¶éãã¼ã¸ + ç¬èªãã¡ã¤ã³",conn5:"ã¡ã¼ã« + Slackï¼5 ã³ãã¯ã¿ï¼",conn88:"SalesforceãHubSpotãShopifyï¼88 以ä¸ï¼",connAll:"å
¨ã³ãã¯ã¿ + ãã©ã¤ãã¼ãéçº",tok500k:"50 ä¸ãã¼ã¯ã³",tok5m:"500 ä¸ãã¼ã¯ã³",yesPrivate:"å«ã + ãã©ã¤ãã¼ãã¨ã³ããã¤ã³ã",sec1:"RLS + TLS 1.3",sec2:"AI ã»ã¼ããã£ã¬ã¤ã¤ã¼ + æ¿èªããã¼",sec3:"SSO/SAMLãVPC/ãªã³ãã¬ã ãç£æ»ãã°",community:"ã³ãã¥ããã£",priority:"åªå
ï¼24 æéï¼",dedicated:"å°ä»»ããã¼ã¸ã£ã¼ + 99.99% SLA"},tiers:{starter:{name:"Starter",desc:"åã㦠AI ã¯ã¼ã¯ãã©ã¼ã¹ãå°å
¥ããå°è¦æ¨¡ãã¼ã åãã"},business:{name:"Business",desc:"AI éç¨ãæ¡å¤§ããæé·ä¼æ¥åãã"},enterprise:{name:"Enterprise",desc:"ããã·ã§ã³ã¯ãªãã£ã«ã«ãªå°å
¥ãè¡ãçµç¹åãã"}},compareTitle:"ãã©ã³æ¯è¼ï¼å
¨é
ç®ï¼",compareSubtitle:"åãã©ã³ã«å«ã¾ããå
容ã並ã¹ã¦æ¯è¼ã",colFeature:"æ©è½",upgradeTitle:"æé·ã«åããã¦ãã¤ã§ãã¢ããã°ã¬ã¼ã",upgradeText:"Starter ããå§ãã¦ãã¯ã³ã¯ãªãã¯ã§ Business ã Enterprise ã¸ãAI 徿¥å¡ã»ãã¼ã¿ã»CRMã»ã¦ã§ããµã¤ãã¯ãã®ã¾ã¾ãä¸éã¨è¨èªã ããæ¡å¼µããã¾ããæ¥å²ãè«æ±ãç§»è¡ä¸è¦ããã¤ã§ãè§£ç´å¯è½ã",upgradeCta:"ã¢ããã°ã¬ã¼ãã®æµããè¦ã",popular:"人æ°",perMonth:"/æ",customPrice:"åå¥è¦ç©ã",ctaStart:"ç¡æãã©ã¤ã¢ã«ãéå§",ctaContact:"å¶æ¥ã«åãåãã",ctaDetail:"ãã©ã³è©³ç´°ãè¦ã",excl:{title:"ãã©ã³ã«å«ã¾ããªãå¤é¨è²»ç¨",intro:"å®å
¨ãªéææ§ï¼Infernex ã®ãã©ã³ã¯ãã©ãããã©ã¼ã ï¼AI 徿¥å¡ãCRMãã¦ã§ããµã¤ããèªååãæ¨è« APIï¼ã対象ã¨ãã¾ãã以ä¸ã®ç¬¬ä¸è
è²»ç¨ã¯ã客æ§ã®ãè² æ
ã§ãã",items:["ãã¡ã¤ã³åã¨å¹´æ¬¡æ´æ°æï¼å¹´éç´ 10ã50 ãã«ï¼â Infernex ã¯ãã¡ã¤ã³ãå販ãã¾ããã","ä»å±ã®ãã®ã使ç¨ããªãå ´åã®å¤é¨ãã¹ãã£ã³ã°ãæ³äººã¡ã¼ã«ã","ææã®ç¬¬ä¸è
ã³ãã¯ã¿ï¼æ¥ç¶ãããã¼ã«ã®å©ç¨æï¼HubSpotãSalesforceãææ SlackãShopifyãLinkedIn Sales Navigator ãªã©ï¼ãå½ç¤¾ã¯æ¥ç¶ãæä¾ããã©ã¤ã»ã³ã¹ã¯å«ã¿ã¾ããã","æã¡è¾¼ã¿ã®ç¬¬ä¸è
API ãã¼ï¼BYOKï¼ã¨ãã®æ¶è²»åã","ãã©ã³æ ãè¶
ãããã¼ã¯ã³è¶
éåï¼å¾é課éï¼ã","ç¾å°ã®ç¨éï¼VATãçºæ¿ææ°æãéè¡ææ°æã","ãªãã·ã§ã³ï¼ç§»è¡æ¯æ´ãã«ã¹ã¿ã ã³ãã¯ã¿éçºãå¼·å SLAã","貴社ãèªå½ã§ç»è¨ãããæ³ä»¤ãéµå®ããäºæ¥ãéå¶ããããã«å¿
è¦ãªå¥ç´ã»ã©ã¤ã»ã³ã¹ã»æ³ç義åï¼åæ¥ç»è¨ãä¿éºãç¾å°ç¨åãGDPR対å¿ãå°éã©ã¤ã»ã³ã¹çï¼ã¯ãã©ã³ã«å«ã¾ãããã客æ§ã®ãè² æ
ã¨ãªãã¾ãã"],note:"ä¼ç¤¾ã®æçµäºç®ã¯ãå復忰ã稼åä¸ã® AI 徿¥å¡æ°ããã¼ã¯ã³æ¶è²»éã«ãã£ã¦å¤åãã¾ãããã©ã³æéã¯ãã©ãããã©ã¼ã ã対象ã¨ããå¤é¨æ¯åºã¯å«ã¿ã¾ããã",disclaimer:"Infernex AI ãã©ãããã©ã¼ã ãéãã¦çæãããè¡çºã»å¤æã»ã³ã³ãã³ãã«ã¤ãã¦ã¯ãã¦ã¼ã¶ã¼ãåç¬ã§è²¬ä»»ãè² ãã¾ããInfernex ã¯ã¡ã¼ã«ã«ãã 24 æé 365 æ¥ã®æè¡ãµãã¼ããæä¾ãã¾ããããµã¼ãã¹ã®ä¸é©åã
811ªå©ç¨ãé©ç¨æ³ä»¤ã¾ãã¯å©ç¨è¦ç´ã«åããå©ç¨ãããã³ãããã«èµ·å ãã忥çã»è²¡åçã»æ³ççµæã«ã¤ãã¦ä¸åã®è²¬ä»»ãè² ãã¾ããã"},home:{badge:"ãã©ã³ã¨æé",title:"ã·ã³ãã«ãªæéãæé¡ 99 ãã«ãã",subtitle:"ä½ãå§ããåã«è²»ç¨ãæç¢ºãStarterãBusinessãEnterpriseããã¤ã§ãã¢ããã°ã¬ã¼ãå¯è½ã",cta:"å
¨ãã©ã³ãæ¯è¼"}}},hi:{wfp:{f:{employees:"AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥",languages:"à¤à¤ªà¤²à¤¬à¥à¤§ à¤à¤¾à¤·à¤¾à¤à¤",website:"à¤à¤à¤ªà¤¨à¥ वà¥à¤¬à¤¸à¤¾à¤à¤ शामिल",crm:"à¤
à¤à¤¤à¤°à¥à¤¨à¤¿à¤¹à¤¿à¤¤ CRM",connectors:"à¤à¤¨à¥à¤à¥à¤à¤° à¤à¤° à¤à¤à¤à¥à¤à¥à¤°à¥à¤¶à¤¨",automations:"à¤à¤à¥à¤®à¥à¤¶à¤¨ à¤à¤° वरà¥à¤à¤«à¤¼à¥à¤²à¥",orchestrator:"मासà¥à¤à¤° AI à¤à¤°à¥à¤à¥à¤¸à¥à¤à¥à¤°à¥à¤à¤°",tokens:"पà¥à¤°à¤¤à¤¿ माह शामिल à¤à¥à¤à¤¨",seats:"मानव सà¥à¤à¥à¤",api:"à¤à¤¨à¥à¤«à¤°à¥à¤à¤¸ API (OpenAI-सà¤à¤à¤¤)",analytics:"ROI डà¥à¤¶à¤¬à¥à¤°à¥à¤¡",security:"सà¥à¤°à¤à¥à¤·à¤¾ à¤à¤° à¤
नà¥à¤ªà¤¾à¤²à¤¨",support:"सहायता"},v:{unlimited:"à¤
सà¥à¤®à¤¿à¤¤",custom:"à¤à¤¸à¥à¤à¤® वà¥à¤²à¥à¤¯à¥à¤®",yes:"शामिल",no:"शामिल नहà¥à¤",limited:"सà¥à¤®à¤¿à¤¤ पहà¥à¤à¤",basic:"बà¥à¤¸à¤¿à¤",advanced:"à¤à¤¡à¤µà¤¾à¤à¤¸à¥à¤¡",full:"पà¥à¤°à¥à¤£ पहà¥à¤à¤",lang1:"1 à¤à¤¾à¤·à¤¾ (à¤à¥à¤°à¤¾à¤¹à¤ à¤à¥ IP सॠसà¥à¤µà¤¤à¤ पहà¤à¤¾à¤¨à¥ à¤à¤)",lang3:"à¤
पनॠपसà¤à¤¦ à¤à¥ 3 à¤à¤¾à¤·à¤¾à¤à¤",lang9:"Infernex à¤à¥ सà¤à¥ à¤à¤¾à¤·à¤¾à¤à¤ (9)",web1:"1-पà¥à¤ साà¤à¤ + बà¥à¤¸à¤¿à¤ SEO",web15:"15 पà¥à¤ तठ+ SEO à¤à¤° hreflang",webUnlimited:"à¤
सà¥à¤®à¤¿à¤¤ पà¥à¤ + à¤à¤¸à¥à¤à¤® डà¥à¤®à¥à¤¨",conn5:"à¤à¤®à¥à¤² + Slack (5 à¤à¤¨à¥à¤à¥à¤à¤°)",conn88:"Salesforce, HubSpot, Shopify (88+)",connAll:"सà¤à¥ à¤à¤¨à¥à¤à¥à¤à¤° + निà¤à¥ à¤à¤¸à¥à¤à¤®",tok500k:"500K à¤à¥à¤à¤¨",tok5m:"5M à¤à¥à¤à¤¨",yesPrivate:"शामिल + निà¤à¥ à¤à¤à¤¡à¤ªà¥à¤à¤à¤",sec1:"RLS + TLS 1.3",sec2:"AI सà¥à¤«à¥à¤à¥ लà¥à¤¯à¤° + à¤
नà¥à¤®à¥à¤¦à¤¨",sec3:"SSO/SAML, VPC/à¤à¤¨-पà¥à¤°à¥à¤®, à¤à¤¡à¤¿à¤ लà¥à¤",community:"à¤à¤®à¥à¤¯à¥à¤¨à¤¿à¤à¥",priority:"पà¥à¤°à¤¾à¤¥à¤®à¤¿à¤à¤¤à¤¾ (24 à¤à¤à¤à¥)",dedicated:"समरà¥à¤ªà¤¿à¤¤ पà¥à¤°à¤¬à¤à¤§à¤ + 99.99% SLA"},tiers:{starter:{name:"Starter",desc:"पहलॠAI वरà¥à¤à¤«à¥à¤°à¥à¤¸ शà¥à¤°à¥ à¤à¤°à¤¨à¥ वालॠà¤à¥à¤à¥ à¤à¥à¤®à¥à¤ à¤à¥ लिà¤à¥¤"},business:{name:"Business",desc:"AI सà¤à¤à¤¾à¤²à¤¨ बढ़ानॠवालॠà¤à¤à¤ªà¤¨à¤¿à¤¯à¥à¤ à¤à¥ लिà¤à¥¤"},enterprise:{name:"Enterprise",desc:"मिशन-à¤à¥à¤°à¤¿à¤à¤¿à¤à¤² तà¥à¤¨à¤¾à¤¤à¥ वालॠसà¤à¤à¤ नà¥à¤ à¤à¥ लिà¤à¥¤"}},compareTitle:"पà¥à¤°à¥à¤£ पà¥à¤²à¤¾à¤¨ तà¥à¤²à¤¨à¤¾",compareSubtitle:"हर पà¥à¤²à¤¾à¤¨ मà¥à¤ शामिल सब à¤à¥à¤, साथ-साथ।",colFeature:"सà¥à¤µà¤¿à¤§à¤¾",upgradeTitle:"à¤à¤¬ à¤à¤ª बढ़à¥à¤, तब à¤
पà¤à¥à¤°à¥à¤¡ à¤à¤°à¥à¤",upgradeText:"Starter सॠशà¥à¤°à¥ à¤à¤°à¥à¤, à¤à¤ à¤à¥à¤²à¤¿à¤ मà¥à¤ Business या Enterprise पर à¤à¤¾à¤à¤à¥¤ à¤à¤ªà¤à¥ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥, डà¥à¤à¤¾, CRM à¤à¤° वà¥à¤¬à¤¸à¤¾à¤à¤ सà¥à¤°à¤à¥à¤·à¤¿à¤¤ रहतॠहà¥à¤ â à¤à¥à¤µà¤² सà¥à¤®à¤¾à¤à¤ à¤à¤° à¤à¤¾à¤·à¤¾à¤à¤ बढ़तॠहà¥à¤à¥¤ पà¥à¤°à¥-रà¥à¤à¤¾ बिलिà¤à¤, à¤à¥à¤ माà¤à¤à¥à¤°à¥à¤¶à¤¨ नहà¥à¤, à¤à¤à¥ à¤à¥ रदà¥à¤¦ à¤à¤°à¥à¤à¥¤",upgradeCta:"à¤
पà¤à¥à¤°à¥à¤¡ पथ दà¥à¤à¥à¤",popular:"सबसॠलà¥à¤à¤ªà¥à¤°à¤¿à¤¯",perMonth:"/माह",customPrice:"à¤à¤¸à¥à¤à¤®",ctaStart:"मà¥à¤«à¤¼à¥à¤¤ à¤à¥à¤°à¤¾à¤¯à¤² शà¥à¤°à¥ à¤à¤°à¥à¤",ctaContact:"सà¥à¤²à¥à¤¸ सॠसà¤à¤ªà¤°à¥à¤ à¤à¤°à¥à¤",ctaDetail:"पà¥à¤²à¤¾à¤¨ à¤à¤¾ पà¥à¤°à¤¾ विवरण दà¥à¤à¥à¤",excl:{title:"à¤à¤ªà¤à¥ पà¥à¤²à¤¾à¤¨ मà¥à¤ शामिल न हà¥à¤¨à¥ वालॠबाहरॠà¤à¤°à¥à¤",intro:"पà¥à¤°à¥ पारदरà¥à¤¶à¤¿à¤¤à¤¾: à¤à¤ªà¤à¤¾ Infernex पà¥à¤²à¤¾à¤¨ पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤® à¤à¥ à¤à¤µà¤° à¤à¤°à¤¤à¤¾ हॠ(AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥, CRM, वà¥à¤¬à¤¸à¤¾à¤à¤, à¤à¤à¥à¤®à¥à¤¶à¤¨, à¤à¤¨à¥à¤«à¤¼à¤°à¥à¤à¤¸ API)। नà¥à¤à¥ दिठà¤à¤ तà¥à¤¤à¥à¤¯-पà¤à¥à¤· à¤à¤°à¥à¤ à¤à¤ªà¤à¥ à¤à¤° सॠरहतॠहà¥à¤à¥¤",items:["डà¥à¤®à¥à¤¨ नाम à¤à¤° à¤à¤¸à¤à¤¾ वारà¥à¤·à¤¿à¤ नवà¥à¤¨à¥à¤à¤°à¤£ (~$10â50/वरà¥à¤·) â Infernex डà¥à¤®à¥à¤¨ नहà¥à¤ बà¥à¤à¤¤à¤¾à¥¤","यदि à¤à¤ª शामिल सà¥à¤µà¤¾ à¤à¤¾ à¤à¤ªà¤¯à¥à¤ नहà¥à¤ à¤à¤°à¤¤à¥ तॠबाहरॠहà¥à¤¸à¥à¤à¤¿à¤à¤ या पà¥à¤°à¥à¤«à¥à¤¶à¤¨à¤² à¤à¤®à¥à¤²à¥¤","à¤à¥à¤à¤¤à¤¾à¤¨ वालॠतà¥à¤¤à¥à¤¯-पà¤à¥à¤· à¤à¤¨à¥à¤à¥à¤à¤°: à¤à¤ªà¤à¥ à¤à¥à¤¡à¤¼à¥ à¤à¤ à¤à¥à¤²à¥à¤¸ à¤à¥ सदसà¥à¤¯à¤¤à¤¾ (HubSpot, Salesforce, paid Slack, Shopify, LinkedIn Sales Navigatorâ¦)। हम à¤à¤¨à¥à¤à¥à¤¶à¤¨ दà¥à¤¤à¥ हà¥à¤, लाà¤à¤¸à¥à¤à¤¸ नहà¥à¤à¥¤","à¤à¤ªà¤à¥ à¤
पनॠतà¥à¤¤à¥à¤¯-पà¤à¥à¤· API à¤à¥à¤à¤à¤¿à¤¯à¤¾à¤ (BYOK) à¤à¤° à¤à¤¨à¤à¤¾ à¤à¤ªà¤¯à¥à¤à¥¤","पà¥à¤²à¤¾à¤¨ à¤à¥à¤à¤¾ सॠà¤
धिठà¤à¥à¤à¤¨ à¤à¤ªà¤¯à¥à¤, à¤à¤ªà¤¯à¥à¤ à¤à¥ à¤à¤§à¤¾à¤° पर बिल।","सà¥à¤¥à¤¾à¤¨à¥à¤¯ à¤à¤°/VAT, मà¥
811दà¥à¤°à¤¾ विनिमय à¤à¤° बà¥à¤à¤ शà¥à¤²à¥à¤à¥¤","वà¥à¤à¤²à¥à¤ªà¤¿à¤ सà¥à¤µà¤¾à¤à¤: सहायठमाà¤à¤à¥à¤°à¥à¤¶à¤¨, à¤à¤¸à¥à¤à¤® à¤à¤¨à¥à¤à¥à¤à¤°, मà¤à¤¬à¥à¤¤ SLA।","à¤à¤ªà¤à¥ à¤à¤à¤ªà¤¨à¥ à¤à¥ à¤à¤ªà¤à¥ दà¥à¤¶ मà¥à¤ पà¤à¤à¥à¤à¥à¤¤, à¤
नà¥à¤ªà¤¾à¤²à¤¨à¤¶à¥à¤² à¤à¤° सà¤à¤à¤¾à¤²à¤¨ यà¥à¤à¥à¤¯ बनानॠà¤à¥ लिठà¤à¤µà¤¶à¥à¤¯à¤ à¤
नà¥à¤¬à¤à¤§, लाà¤à¤¸à¥à¤à¤¸ à¤à¤° à¤à¤¾à¤¨à¥à¤¨à¥ दायितà¥à¤µ (à¤à¤à¤ªà¤¨à¥ पà¤à¤à¥à¤à¤°à¤£, बà¥à¤®à¤¾, सà¥à¤¥à¤¾à¤¨à¥à¤¯ à¤à¤°, GDPR à¤
नà¥à¤ªà¤¾à¤²à¤¨, पà¥à¤¶à¥à¤µà¤° लाà¤à¤¸à¥à¤à¤¸ à¤à¤¦à¤¿) यà¥à¤à¤¨à¤¾ मà¥à¤ शामिल नहà¥à¤ हà¥à¤ à¤à¤° à¤à¤ªà¤à¥ à¤à¤°à¥à¤ पर रहतॠहà¥à¤à¥¤"],note:"à¤à¤ªà¤à¥ à¤à¤à¤ªà¤¨à¥ à¤à¤¾ à¤
à¤à¤¤à¤¿à¤® बà¤à¤ à¤à¤ªà¤à¥ पà¥à¤¨à¤°à¤¾à¤µà¥à¤¤à¥à¤¤à¤¿à¤¯à¥à¤ à¤à¥ सà¤à¤à¥à¤¯à¤¾, सà¤à¥à¤°à¤¿à¤¯ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ à¤à¤° à¤à¥à¤à¤¨ à¤à¤ªà¤¤ पर निरà¥à¤à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤ पà¥à¤²à¤¾à¤¨ à¤à¥ à¤à¥à¤®à¤¤ पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤® à¤à¥ à¤à¤µà¤° à¤à¤°à¤¤à¥ हà¥, à¤à¤ªà¤à¥ बाहरॠà¤à¤°à¥à¤ à¤à¥ नहà¥à¤à¥¤",disclaimer:"Infernex AI पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤® à¤à¥ माधà¥à¤¯à¤® सॠà¤à¤¿à¤ à¤à¤ à¤à¤¾à¤°à¥à¤¯à¥à¤, निरà¥à¤£à¤¯à¥à¤ à¤à¤° à¤à¤¤à¥à¤ªà¤¨à¥à¤¨ सामà¤à¥à¤°à¥ à¤à¥ लिठपà¥à¤°à¥à¤£ रà¥à¤ª सॠà¤à¤ªà¤¯à¥à¤à¤à¤°à¥à¤¤à¤¾ हॠà¤à¤¤à¥à¤¤à¤°à¤¦à¤¾à¤¯à¥ हà¥à¥¤ Infernex à¤à¤®à¥à¤² दà¥à¤µà¤¾à¤°à¤¾ 24/7 तà¤à¤¨à¥à¤à¥ सहायता पà¥à¤°à¤¦à¤¾à¤¨ à¤à¤°à¤¤à¤¾ हà¥, परà¤à¤¤à¥ सà¥à¤µà¤¾ à¤à¥ दà¥à¤°à¥à¤ªà¤¯à¥à¤, लाà¤à¥ à¤à¤¾à¤¨à¥à¤¨à¥à¤ या हमारॠà¤à¤ªà¤¯à¥à¤ शरà¥à¤¤à¥à¤ à¤à¥ विरà¥à¤¦à¥à¤§ à¤à¤ªà¤¯à¥à¤, तथा à¤à¤¸à¤¸à¥ à¤à¤¤à¥à¤ªà¤¨à¥à¤¨ वà¥à¤¯à¤¾à¤µà¤¸à¤¾à¤¯à¤¿à¤, वितà¥à¤¤à¥à¤¯ या à¤à¤¾à¤¨à¥à¤¨à¥ परिणामà¥à¤ à¤à¥ लिठà¤à¤¿à¤¸à¥ à¤à¥ दायितà¥à¤µ सॠà¤à¤¨à¤à¤¾à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤"},home:{badge:"पà¥à¤²à¤¾à¤¨ à¤à¤° मà¥à¤²à¥à¤¯",title:"सरल मà¥à¤²à¥à¤¯, $99/माह सà¥",subtitle:"बनानॠसॠपहलॠà¤à¤¾à¤¨à¥à¤ à¤à¤¿ à¤à¤ª à¤à¤¿à¤¤à¤¨à¤¾ à¤à¥à¤à¤¤à¤¾à¤¨ à¤à¤°à¥à¤à¤à¥à¥¤ Starter, Business या Enterprise â à¤à¤à¥ à¤à¥ à¤
पà¤à¥à¤°à¥à¤¡ à¤à¤°à¥à¤à¥¤",cta:"सà¤à¥ पà¥à¤²à¤¾à¤¨ à¤à¥ तà¥à¤²à¤¨à¤¾ à¤à¤°à¥à¤"}}},ar:{wfp:{f:{employees:"Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",languages:"اÙÙØºØ§Øª اÙÙ
ØªØ§ØØ©",website:"Ù
ÙÙØ¹ Ø§ÙØ´Ø±ÙØ© Ù
Ø´Ù
ÙÙ",crm:"ÙØ¸Ø§Ù
CRM Ù
دÙ
ج",connectors:"اÙÙ
ÙØµÙات ÙØ§ÙØªÙØ§Ù
ÙØ§Øª",automations:"Ø§ÙØ£ØªÙ
تة ÙØ³Ùر Ø§ÙØ¹Ù
Ù",orchestrator:"ÙÙÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù Ø§ÙØ±Ø¦ÙسÙ",tokens:"Ø§ÙØ±Ù
ÙØ² اÙÙ
Ø´Ù
ÙÙØ© / Ø´ÙØ±ÙÙØ§",seats:"اÙÙ
ÙØ§Ø¹Ø¯ Ø§ÙØ¨Ø´Ø±ÙØ©",api:"ÙØ§Ø¬ÙØ© Ø§ÙØ§Ø³ØªØ¯Ùا٠(Ù
ØªÙØ§ÙÙØ© Ù
ع OpenAI)",analytics:"ÙÙØØ© عائد Ø§ÙØ§Ø³ØªØ«Ù
ار",security:"Ø§ÙØ£Ù
Ø§Ù ÙØ§ÙاÙ
تثاÙ",support:"Ø§ÙØ¯Ø¹Ù
"},v:{unlimited:"ØºÙØ± Ù
ØØ¯Ùد",custom:"ØØ¬Ù
Ù
خصص",yes:"Ù
Ø´Ù
ÙÙ",no:"ØºÙØ± Ù
Ø´Ù
ÙÙ",limited:"ÙØµÙÙ Ù
ØØ¯Ùد",basic:"أساسÙ",advanced:"Ù
ØªÙØ¯Ù
",full:"ÙØµÙÙ ÙØ§Ù
Ù",lang1:"ÙØºØ© ÙØ§ØØ¯Ø© (تÙÙØªØ´Ù تÙÙØ§Ø¦ÙÙØ§ Ù
٠عÙÙØ§Ù IP ÙÙØ¹Ù
ÙÙ)",lang3:"3 ÙØºØ§Øª Ù
Ù Ø§Ø®ØªÙØ§Ø±Ù",lang9:"جÙ
ÙØ¹ ÙØºØ§Øª Infernex (9)",web1:"Ù
ÙÙØ¹ Ù
Ù ØµÙØØ© ÙØ§ØØ¯Ø© + SEO أساسÙ",web15:"ØØªÙ 15 ØµÙØØ© + SEO Ù hreflang",webUnlimited:"ØµÙØØ§Øª ØºÙØ± Ù
ØØ¯Ùدة + ÙØ·Ø§Ùات Ù
خصصة",conn5:"Ø§ÙØ¨Ø±Ùد + Slack (5 Ù
ÙØµÙات)",conn88:"Salesforce ÙHubSpot ÙShopify (88+)",connAll:"جÙ
ÙØ¹ اÙÙ
ÙØµÙات + Ù
ÙØµÙات خاصة",tok500k:"500 Ø£Ù٠رÙ
ز",tok5m:"5 Ù
ÙØ§ÙÙ٠رÙ
ز",yesPrivate:"Ù
Ø´Ù
ÙÙ + ÙÙØ§Ø· ÙÙØ§ÙØ© خاصة",sec1:"RLS + TLS 1.3",sec2:"Ø·Ø¨ÙØ© Ø£Ù
Ø§Ù Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·Ùاع٠+ اÙÙ
ÙØ§ÙÙØ§Øª",sec3:"SSO/SAMLØ VPC/Ù
ØÙÙØ Ø³Ø¬ÙØ§Øª Ø§ÙØªØ¯ÙÙÙ",community:"اÙÙ
جتÙ
ع",priority:"Ø£ÙÙÙÙØ© (24 ساعة)",dedicated:"Ù
Ø¯ÙØ± Ù
خصص + Ø§ØªÙØ§ÙÙØ© 99.99%"},tiers:{starter:{name:"Starter",desc:"ÙÙÙØ±Ù Ø§ÙØµØºÙرة Ø§ÙØªÙ تطÙ٠أÙÙ ÙÙØ© عÙ
٠ذÙÙØ©."},business:{name:"Business",desc:"ÙÙØ´Ø±Ùات اÙÙØ§Ù
ÙØ© Ø§ÙØªÙ ØªÙØ³Ùع عÙ
ÙÙØ§Øª Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ."},enterprise:{name:"Enterprise",desc:"ÙÙÙ
ؤسسات ذات عÙ
ÙÙØ§Øª اÙÙØ´Ø± Ø§ÙØØ±Ø¬Ø©."}},compareTitle:"Ù
ÙØ§Ø±ÙØ© ÙØ§Ù
ÙØ© ÙÙØ®Ø·Ø·",compareSubtitle:"ÙÙ Ù
ا تتضÙ
ÙÙ Ù٠خطة Ø¬ÙØ¨Ùا Ø¥ÙÙ Ø¬ÙØ¨.",colFeature:"اÙÙ
ÙØ²Ø©",upgradeTitle:"ارتÙÙ ÙÙÙ
ا ÙÙ
ÙÙØª",upgradeText:"ابدأ بخطة Starter ÙØ§ÙتÙ٠إÙÙ Business Ø£Ù Enterprise بÙÙØ±Ø© ÙØ§ØØ¯Ø©. ÙØªÙ
Ø§ÙØ§ØØªÙاظ بÙ
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù ÙØ§ÙØ¨ÙØ§Ùات ÙCRM ÙØ§ÙÙ
ÙÙØ¹ â ØªØªÙسع Ø§ÙØØ¯ÙØ¯ ÙØ§ÙÙØºØ§Øª ÙÙØ·. ÙÙØªØ±Ø© ØªÙØ§Ø³Ø¨ÙØ©Ø Ø¨ÙØ§ ترØÙÙØ ÙØ¥Ùغاء Ù٠أ٠ÙÙØª.",upgradeCta:"عرض Ù
سار Ø§ÙØªØ±ÙÙØ©",popular:"Ø§ÙØ£Ùثر Ø§Ø®ØªÙØ§Ø±Ùا",perMonth:"/Ø´ÙØ±ÙÙØ§",customPrice:"ØØ³Ø¨ Ø§ÙØ·Ùب",ctaStart:"ابدأ Ø§ÙØªØ¬Ø±Ø¨Ø© اÙÙ
جاÙÙØ©",ctaContact:"ØªÙØ§ØµÙ Ù
ع اÙÙ
Ø¨ÙØ¹Ø§Øª",ctaDetail:"عرض ØªÙØ§ØµÙÙ Ø§ÙØ®Ø·Ø©",excl:{title:"ØªÙØ§ÙÙÙ Ø®Ø§Ø±Ø¬ÙØ© ØºÙØ± Ù
Ø´Ù
ÙÙØ© Ù٠خطتÙ",intro:"Ø´ÙØ§ÙÙØ© ÙØ§Ù
ÙØ©: تغط٠خطة Infernex اÙÙ
ÙØµØ© (Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹ÙØ CRMØ Ø§ÙÙ
ÙÙØ¹Ø Ø§ÙØ£ØªÙ
ØªØ©Ø ÙØ§Ø¬ÙØ© Ø§ÙØ§Ø³ØªØ¯ÙاÙ). Ø§ÙØªÙاÙÙÙ Ø§ÙØ®Ø§Ø±Ø¬ÙØ© Ø§ÙØªØ§ÙÙØ© تبÙ٠عÙ٠عاتÙÙ.",items:["اسÙ
اÙÙØ·Ø§Ù ÙØªØ¬Ø¯ÙØ¯Ù Ø§ÙØ³ÙÙÙ (~10â50 دÙÙØ§Ø±Ùا/Ø³ÙØ©) â Infernex ÙØ§ ØªØ¨ÙØ¹ اÙÙØ·Ø§Ùات.","Ø§ÙØ§Ø³ØªØ¶Ø§ÙØ© Ø§ÙØ®Ø§Ø±Ø¬ÙØ© Ø£Ù Ø§ÙØ¨Ø±Ùد اÙÙ
ÙÙ٠إذا ÙÙ
تستخدÙ
اÙÙ
تضÙ
ÙÙ.","اÙÙ
ÙØµÙÙØ§Øª اÙÙ
دÙÙØ¹Ø© Ù
Ù Ø£Ø·Ø±Ø§Ù Ø«Ø§ÙØ«Ø©: Ø§Ø´ØªØ±Ø§ÙØ§Øª Ø§ÙØ£Ø¯Ùات Ø§ÙØªÙ ØªØ±Ø¨Ø·ÙØ§ (HubSpotØ SalesforceØ Slack اÙÙ
دÙÙØ¹Ø ShopifyØ LinkedIn Sales Navigatorâ¦). ÙØÙ ÙÙÙÙØ± Ø§ÙØ±Ø¨Ø· ÙØ§ Ø§ÙØªØ±Ø®Ùص.","Ù
ÙØ§ØªÙØ ÙØ§Ø¬Ùات برÙ
جة Ø®Ø§Ø±Ø¬ÙØ© ØªØ¬ÙØ¨Ùا (BYOK) ÙØ§Ø³ØªÙÙØ§ÙÙØ§.","ØªØ¬Ø§ÙØ² Ø§ÙØ±Ù
ÙØ² (tokens) ÙÙÙ ØØµØ© Ø§ÙØ®Ø·Ø©Ø ÙÙÙÙÙØªØ± ØØ³Ø¨ Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù
.","Ø§ÙØ¶Ø±Ø§Ø¦Ø¨/اÙÙÙÙ
Ø© اÙÙ
Ø¶Ø§ÙØ© اÙÙ
ØÙÙØ© ÙØ±Ø³ÙÙ
Ø§ÙØµØ±Ù ÙØ§ÙرسÙÙ
Ø§ÙØ¨ÙÙÙØ©.","خدÙ
ات Ø§Ø®ØªÙØ§Ø±ÙØ©: ترØÙÙ Ù
ÙØ³Ø§Ø¹Ø¯Ø Ù
ÙØµÙÙ Ù
Ø®ØµØµØ Ø§ØªÙØ§ÙÙØ© Ù
ستÙ٠خدÙ
Ø© Ù
عززة.","Ø§ÙØ¹ÙÙØ¯ ÙØ§ÙØªØ±Ø§Ø®ÙØµ ÙØ§ÙØ§ÙØªØ²Ø§Ù
ات اÙÙØ§ÙÙÙÙØ© اÙÙØ§Ø²Ù
Ø© ÙØªØ³Ø¬ÙÙ Ø´Ø±ÙØªÙ ÙØ§Ù
تثاÙÙØ§ ÙØªØ´ØºÙÙÙØ§ ÙÙ Ø¨ÙØ¯Ù (Ø§ÙØ³Ø¬Ù Ø§ÙØªØ¬Ø§Ø±ÙØ Ø§ÙØªØ£Ù
ÙÙØ§ØªØ Ø§ÙØ¶Ø±Ø§Ø¦Ø¨ اÙÙ
ØÙÙØ©Ø Ø§ÙØ§Ù
تثا٠ÙÙØ§Ø¦ØØ© Ø§ÙØ¹Ø§Ù
Ø© ÙØÙ
Ø§ÙØ© Ø§ÙØ¨ÙØ§ÙØ§ØªØ Ø§ÙØªØ±Ø§Ø®Ùص اÙÙ
ÙÙÙØ©Ø Ø¥ÙØ®) ØºÙØ± Ù
Ø´Ù
ÙÙØ© Ø¨Ø§ÙØ®Ø·Ø© ÙØªØ¨Ù٠عÙÙ ÙÙÙØªÙ."],note:"تعتÙ
د اÙÙ
ÙØ²Ø§ÙÙØ© اÙÙÙØ§Ø¦ÙØ© ÙØ´Ø±Ùت٠عÙ٠عدد Ø§ÙØªÙرارات ÙØ¹Ø¯Ø¯ Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·Ùاع٠اÙÙØ´Ø·ÙÙ ÙØ§Ø³ØªÙÙØ§Ù Ø§ÙØ±Ù
ÙØ². سعر Ø§ÙØ®Ø·Ø© ÙØºØ·Ù اÙÙ
ÙØµØ© ÙÙÙØ³ ÙÙÙØ§ØªÙ Ø§ÙØ®Ø§Ø±Ø¬ÙØ©.",disclaimer:"ÙØªØÙ
Ù٠اÙÙ
ستخدÙ
ÙØØ¯Ù Ø§ÙÙ
سؤÙÙÙØ© Ø¹Ù Ø§ÙØ¥Ø¬Ø±Ø§Ø¡Ø§Øª ÙØ§ÙÙØ±Ø§Ø±Ø§Øª ÙØ§ÙÙ
ØØªÙ٠اÙÙØ§ØªØ¬ عبر Ù
ÙØµØ© Infernex AI. تÙÙÙØ± Infernex دعÙ
ا٠تÙÙÙØ§Ù عÙÙ Ù
دار Ø§ÙØ³Ø§Ø¹Ø© Ø·ÙØ§Ù Ø£ÙØ§Ù
Ø§ÙØ£Ø³Ø¨Ùع عبر Ø§ÙØ¨Ø±Ùد Ø§ÙØ¥ÙÙØªØ±ÙÙÙØ ÙÙÙÙØ§ ØªÙØ®ÙÙ Ù
سؤÙÙÙØªÙا Ø¹Ù Ø£Ù Ø³ÙØ¡ استخداÙ
ÙÙØ®Ø¯Ù
Ø© أ٠استخداÙ
ÙØ§ ÙØªÙاÙÙ Ù
ع اÙÙÙØ§ÙÙ٠اÙÙ
عÙ
ÙÙ Ø¨ÙØ§ Ø£Ù Ø´Ø±ÙØ· Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù
Ø ÙØ¹Ù Ø£Ù Ø¹ÙØ§Ùب ØªØ¬Ø§Ø±ÙØ© Ø£Ù Ù
اÙÙØ© Ø£Ù ÙØ§ÙÙÙÙØ© ØªÙØ¬Ù
ع٠ذÙÙ."},home:{badge:"Ø§ÙØ®Ø·Ø· ÙØ§Ùأسعار",title:"ØªØ³Ø¹ÙØ± Ø¨Ø³ÙØ· ÙØ¨Ø¯Ø£ Ù
Ù 99 دÙÙØ§Ø±Ùا Ø´ÙØ±ÙÙØ§",subtitle:"Ø§Ø¹Ø±Ù Ø¨Ø§ÙØ¶Ø¨Ø· Ù
ا Ø³ØªØ¯ÙØ¹Ù ÙØ¨Ù Ø§ÙØ¨Ùاء. Starter Ø£Ù Business Ø£Ù Enterprise â Ø§ÙترÙÙØ© Ù
ØªØ§ØØ© Ù٠أ٠ÙÙØª.",cta:"ÙØ§Ø±Ù جÙ
ÙØ¹ Ø§ÙØ®Ø·Ø·"}}}},Lr={en:{wfpd:{seoTitle:"{{plan}} plan â AI Workforce pricing & features",seoDesc:"Full detail of the Infernex {{plan}} plan: AI employees, languages, website, CRM, connectors, tokens, security and support.",eyebrow:"PLAN DETAIL",back:"All plans",intro:"Everything you get with {{plan}}, point by point â no surprises.",includedTitle:"What you get, point by point",notIncludedTitle:"What is not included",notIncludedText:"Available on a higher plan â upgrade anytime, your data follows you.",ctaTitle:"Ready to deploy your AI workforce?",ctaText:"Create your company, hire your first AI employees and go live in minutes.",ctaCompare:"Compare with other plans",trust1:"30-day money-back guarantee",trust2:"No credit card required",trust3:"Cancel anytime",faqTitle:"Frequently asked questions",faq:{q1:"Can I change plan later?",a1:"Yes. Upgrade or downgrade in one click; billing is prorated and your AI employees, data, CRM and website are preserved.",q2:"What if I exceed my tokens?",a2:"Extra usage switches to pay-as-you-go with a monthly hard cap you control. Nothing is billed beyond your limit.",q3:"Do I need technical skills?",a3:"No. You describe your business and the Master Orchestrator hires and configures the right AI employees.",q4:"Is my data isolated?",a4:"Yes. Every company is isolated at database level with row-level security. Enterprise adds SSO/SAML, VPC/on-prem and audit logs.",q5:"Are there costs outside the plan?",a5:"Yes, and we state them upfront: domain name, external hosting or pro email, the paid subscriptions of the third-party connectors you plug in, your own API keys (BYOK), token overage, local taxes and optional services. Your final budget also depen
811ds on your number of iterations and token consumption. Contracts, licences and legal obligations required for your company to be registered, compliant and operational in your country (company registration, insurance, local taxes, GDPR compliance, professional licences, etc.) are also not included and remain at your expense.",q6:"Who is responsible for what is done on the platform?",a6:"The user is solely responsible for the actions, decisions and content generated through the Infernex AI platform. Infernex provides 24/7 technical support by email but disclaims any liability for misuse of the service, any use that does not comply with applicable laws or our terms of use, and for the commercial, financial or legal consequences that may result."},d:{employees:"Autonomous AI agents you hire, each with its own role, memory and objectives.",languages:"Languages your AI employees, website and customer replies are delivered in.",website:"A company website generated, hosted and kept up to date by your AI workforce.",crm:"Contacts, deals and pipeline handled automatically by your AI employees.",connectors:"Third-party tools your AI employees can read from and act on.",automations:"Recurring workflows your employees run 24/7 without supervision.",orchestrator:"The master agent that assigns work, arbitrates priorities and supervises every employee.",tokens:"Monthly AI compute included. Extra usage is pay-as-you-go with a hard cap.",seats:"Human team members who can log in and manage the workforce.",api:"OpenAI-compatible endpoint to call Infernex inference from your own code.",analytics:"Hours saved, cost per task and ROI measured in real time.",security:"Data isolation, access controls and compliance guarantees.",support:"How fast you get a human answer when it matters."}}},fr:{wfpd:{seoTitle:"Plan {{plan}} â tarifs et fonctionnalités AI Workforce",seoDesc:"Détail complet du plan Infernex {{plan}} : employés IA, langues, site web, CRM, connecteurs, jetons, sécurité et support.",eyebrow:"DÃTAIL DU PLAN",back:"Tous les plans",intro:"Tout ce que vous obtenez avec {{plan}}, point par point â sans surprise.",includedTitle:"Ce que vous obtenez, point par point",notIncludedTitle:"Ce qui n'est pas inclus",notIncludedText:"Disponible sur un plan supérieur â évoluez quand vous voulez, vos données vous suivent.",ctaTitle:"Prêt à déployer votre workforce IA ?",ctaText:"Créez votre entreprise, recrutez vos premiers employés IA et lancez-vous en quelques minutes.",ctaCompare:"Comparer avec les autres plans",trust1:"Garantie satisfait ou remboursé 30 jours",trust2:"Sans carte bancaire",trust3:"Annulable à tout moment",faqTitle:"Questions fréquentes",faq:{q1:"Puis-je changer de plan plus tard ?",a1:"Oui. Montée ou descente de gamme en un clic ; facturation au prorata, et vos employés IA, données, CRM et site web sont conservés.",q2:"Que se passe-t-il si je dépasse mes jetons ?",a2:"L'usage supplémentaire bascule en paiement à l'usage avec un plafond mensuel que vous contrôlez. Rien n'est facturé au-delà de votre limite.",q3:"Faut-il des compétences techniques ?",a3:"Non. Vous décrivez votre activité et l'Agent IA Maître recrute et configure les bons employés IA.",q4:"Mes données sont-elles isolées ?",a4:"Oui. Chaque entreprise est isolée au niveau base de données (row-level security). Enterprise ajoute SSO/SAML, VPC/on-prem et journaux d'audit.",q5:"Y a-t-il des frais en dehors du plan ?",a5:"Oui, et nous les annonçons clairement : nom de domaine, hébergement ou e-mail pro externe, abonnements payants des connecteurs tiers que vous branchez, vos propres clés API (BYOK), dépassement de jetons, taxes locales et prestations optionnelles. Le budget final dépend aussi de votre nombre d'itérations et de votre consommation de jetons. Les contrats, licences et obligations légales obligatoires pour que votre entreprise soit immatriculée, conforme et opérationnelle dans votre pays (registre du commerce, assurances, fiscalité locale, RGPD, licences professionnelles, etc.) ne sont pas non plus inclus et restent à votre charge.",q6:"Qui est responsable de ce qui est fait sur la plateforme ?",a6:"L'utilisateur est seul responsable des actions, décisions et contenus générés via la plateforme Infernex AI. Infernex fournit une assistance technique 24h/24 et 7j/7 par e-mail, mais décline toute responsabilité en cas de mauvaise utilisation du service, d'usage non conforme aux lois applicables ou à nos conditions d'utilisation, ainsi que des conséquences commerciales, financières ou légales qui en découleraie
811nt."},d:{employees:"Agents IA autonomes que vous recrutez, chacun avec son rôle, sa mémoire et ses objectifs.",languages:"Langues dans lesquelles vos employés IA, votre site et vos réponses clients sont livrés.",website:"Un site d'entreprise généré, hébergé et mis à jour en continu par votre workforce IA.",crm:"Contacts, opportunités et pipeline gérés automatiquement par vos employés IA.",connectors:"Outils tiers que vos employés IA peuvent lire et piloter.",automations:"Workflows récurrents exécutés 24/7 sans supervision.",orchestrator:"L'agent maître qui répartit le travail, arbitre les priorités et supervise chaque employé.",tokens:"Puissance IA incluse chaque mois. Au-delà : paiement à l'usage avec plafond strict.",seats:"Membres humains pouvant se connecter et piloter la workforce.",api:"Endpoint compatible OpenAI pour appeler l'inférence Infernex depuis votre code.",analytics:"Heures économisées, coût par tâche et ROI mesurés en temps réel.",security:"Isolation des données, contrôles d'accès et garanties de conformité.",support:"La rapidité avec laquelle un humain vous répond quand ça compte."}}},es:{wfpd:{seoTitle:"Plan {{plan}} â precios y funciones de AI Workforce",seoDesc:"Detalle completo del plan Infernex {{plan}}: empleados IA, idiomas, sitio web, CRM, conectores, tokens, seguridad y soporte.",eyebrow:"DETALLE DEL PLAN",back:"Todos los planes",intro:"Todo lo que obtienes con {{plan}}, punto por punto â sin sorpresas.",includedTitle:"Lo que obtienes, punto por punto",notIncludedTitle:"Lo que no está incluido",notIncludedText:"Disponible en un plan superior â mejora cuando quieras, tus datos te siguen.",ctaTitle:"¿Listo para desplegar tu plantilla de IA?",ctaText:"Crea tu empresa, contrata a tus primeros empleados IA y lánzate en minutos.",ctaCompare:"Comparar con otros planes",trust1:"GarantÃa de devolución de 30 dÃas",trust2:"Sin tarjeta de crédito",trust3:"Cancela cuando quieras",faqTitle:"Preguntas frecuentes",faq:{q1:"¿Puedo cambiar de plan más adelante?",a1:"SÃ. Sube o baja de plan en un clic; la facturación es prorrateada y se conservan tus empleados IA, datos, CRM y sitio web.",q2:"¿Qué pasa si supero mis tokens?",a2:"El uso adicional pasa a pago por uso con un lÃmite mensual que tú controlas. Nada se factura por encima de tu lÃmite.",q3:"¿Necesito conocimientos técnicos?",a3:"No. Describes tu negocio y el Orquestador Maestro contrata y configura a los empleados IA adecuados.",q4:"¿Están aislados mis datos?",a4:"SÃ. Cada empresa está aislada a nivel de base de datos con row-level security. Enterprise añade SSO/SAML, VPC/on-prem y registros de auditorÃa.",q5:"¿Hay costes fuera del plan?",a5:"SÃ, y los indicamos claramente: nombre de dominio, alojamiento o correo profesional externo, suscripciones de pago de los conectores de terceros que conectes, tus propias claves API (BYOK), exceso de tokens, impuestos locales y servicios opcionales. El presupuesto final depende también de tus iteraciones y consumo de tokens. Los contratos, licencias y obligaciones legales necesarios para que su empresa esté registrada, en conformidad y operativa en su paÃs (registro mercantil, seguros, fiscalidad local, RGPD, licencias profesionales, etc.) tampoco están incluidos y corren por su cuenta.",q6:"¿Quién es responsable de lo que se hace en la plataforma?",a6:"El usuario es el único responsable de las acciones, decisiones y contenidos generados a través de la plataforma Infernex AI. Infernex ofrece soporte técnico 24/7 por correo electrónico, pero declina toda responsabilidad por un uso indebido del servicio, por un uso no conforme con las leyes aplicables o con nuestras condiciones de uso, asà como por las consecuencias comerciales, financieras o legales que puedan derivarse."},d:{employees:"Agentes IA autónomos que contratas, cada uno con su rol, memoria y objetivos.",languages:"Idiomas en los que se entregan tus empleados IA, tu web y las respuestas a clientes.",website:"Un sitio web corporativo generado, alojado y actualizado por tu plantilla de IA.",crm:"Contactos, op
811ortunidades y pipeline gestionados automáticamente por tus empleados IA.",connectors:"Herramientas de terceros que tus empleados IA pueden leer y accionar.",automations:"Flujos recurrentes que tus empleados ejecutan 24/7 sin supervisión.",orchestrator:"El agente maestro que asigna el trabajo, arbitra prioridades y supervisa a cada empleado.",tokens:"Cómputo IA incluido cada mes. El exceso es pago por uso con lÃmite estricto.",seats:"Miembros humanos que pueden acceder y gestionar la plantilla.",api:"Endpoint compatible con OpenAI para llamar a la inferencia de Infernex desde tu código.",analytics:"Horas ahorradas, coste por tarea y ROI medidos en tiempo real.",security:"Aislamiento de datos, controles de acceso y garantÃas de cumplimiento.",support:"La rapidez con la que un humano te responde cuando importa."}}},de:{wfpd:{seoTitle:"{{plan}}-Tarif â Preise und Funktionen der AI Workforce",seoDesc:"Vollständige Details zum Infernex-Tarif {{plan}}: KI-Mitarbeiter, Sprachen, Website, CRM, Konnektoren, Tokens, Sicherheit und Support.",eyebrow:"TARIFDETAILS",back:"Alle Tarife",intro:"Alles, was Sie mit {{plan}} erhalten â Punkt für Punkt, ohne Ãberraschungen.",includedTitle:"Das erhalten Sie, Punkt für Punkt",notIncludedTitle:"Nicht enthalten",notIncludedText:"In einem höheren Tarif verfügbar â jederzeit upgraden, Ihre Daten wandern mit.",ctaTitle:"Bereit, Ihre KI-Belegschaft zu starten?",ctaText:"Firma anlegen, erste KI-Mitarbeiter einstellen und in wenigen Minuten live gehen.",ctaCompare:"Mit anderen Tarifen vergleichen",trust1:"30 Tage Geld-zurück-Garantie",trust2:"Keine Kreditkarte nötig",trust3:"Jederzeit kündbar",faqTitle:"Häufige Fragen",faq:{q1:"Kann ich später den Tarif wechseln?",a1:"Ja. Up- oder Downgrade mit einem Klick; anteilige Abrechnung, und KI-Mitarbeiter, Daten, CRM und Website bleiben erhalten.",q2:"Was passiert bei Ãberschreitung der Tokens?",a2:"Zusätzliche Nutzung läuft als Pay-as-you-go mit einem monatlichen Hard Cap, den Sie festlegen. Ãber Ihr Limit hinaus wird nichts berechnet.",q3:"Brauche ich technische Kenntnisse?",a3:"Nein. Sie beschreiben Ihr Geschäft, der Master-Orchestrator stellt die passenden KI-Mitarbeiter ein und konfiguriert sie.",q4:"Sind meine Daten isoliert?",a4:"Ja. Jede Firma ist auf Datenbankebene per Row-Level-Security isoliert. Enterprise ergänzt SSO/SAML, VPC/on-prem und Audit-Logs.",q5:"Gibt es Kosten auÃerhalb des Tarifs?",a5:"Ja, und wir nennen sie offen: Domainname, externes Hosting oder Business-E-Mail, kostenpflichtige Abos der angebundenen Drittanbieter-Konnektoren, eigene API-Schlüssel (BYOK), Token-Ãberschreitung, lokale Steuern und optionale Leistungen. Das Endbudget hängt zudem von Ihren Iterationen und Ihrem Token-Verbrauch ab. Auch Verträge, Lizenzen und rechtliche Pflichten, die erforderlich sind, damit Ihr Unternehmen in Ihrem Land eingetragen, konform und betriebsbereit ist (Handelsregister, Versicherungen, lokale Steuern, DSGVO, Berufslizenzen usw.), sind nicht enthalten und gehen zu Ihren Lasten.",q6:"Wer ist für das verantwortlich, was auf der Plattform geschieht?",a6:"Der Nutzer ist allein verantwortlich für die über die Infernex-AI-Plattform erzeugten Handlungen, Entscheidungen und Inhalte. Infernex bietet technischen Support rund um die Uhr per E-Mail, lehnt jedoch jede Haftung für missbräuchliche Nutzung des Dienstes, für eine nicht gesetzeskonforme oder gegen unsere Nutzungsbedingungen verstoÃende Nutzung sowie für daraus resultierende geschäftliche, finanzielle oder rechtliche Folgen ab."},d:{employees:"Autonome KI-Agenten, die Sie einstellen â jeder mit eigener Rolle, Gedächtnis und Zielen.",languages:"Sprachen, in denen KI-Mitarbeiter, Website und Kundenantworten ausgeliefert werden.",website:"Eine Unternehmenswebsite, erzeugt, gehostet und laufend aktualisiert von Ihrer KI-Belegschaft.",crm:"Kontakte, Deals und Pipeline werden automatisch von Ihren KI-Mitarbeitern gepflegt.",connectors:"Drittsysteme, die Ihre KI-Mitarbeiter lesen und steuern können.",automations:"Wiederkehrende Workflows, die rund um die Uhr ohne Aufsicht laufen.",orchestrator:"Der Master-Agent verteilt Aufgaben, priorisiert und überwacht jeden Mitarbeiter.",tokens:"Monatlich enthaltene KI-Rechenleistung. Mehrverbrauch als Pay-as-you-go mit Hard Cap.",seats:"Menschliche Teammitglieder mit Zugang zur Steuerung der Belegschaft.",api:"OpenAI-kompatibler Endpunkt für Infernex-Inferenz aus Ihrem eigenen Code.",analytics:"Eingesparte Stunden, Kosten pro Aufgabe und ROI in Echtzeit gemessen.",security:"Datenisolation, Zugriffskontrollen und Compliance-Garantien.",support:"Wie schnell Sie eine menschliche Antwort erhalten, wenn es zählt."}}},pt:{wfpd:{seoTitle:"Plano {{plan}} â preços e recursos do AI Workforce",seoDesc:"Detalhe completo do plano Infernex {{plan}}: funcionários IA, idiomas, site, CRM, conectores, tokens, segurança e suporte.",eyebrow:"DETALHE DO PLANO",back:"Todos os planos",intro:"Tudo o que você recebe c
811om {{plan}}, ponto a ponto â sem surpresas.",includedTitle:"O que você recebe, ponto a ponto",notIncludedTitle:"O que não está incluÃdo",notIncludedText:"DisponÃvel num plano superior â faça upgrade quando quiser, os seus dados acompanham.",ctaTitle:"Pronto para lançar a sua força de trabalho IA?",ctaText:"Crie a sua empresa, contrate os primeiros funcionários IA e entre no ar em minutos.",ctaCompare:"Comparar com outros planos",trust1:"Garantia de reembolso em 30 dias",trust2:"Sem cartão de crédito",trust3:"Cancele quando quiser",faqTitle:"Perguntas frequentes",faq:{q1:"Posso mudar de plano depois?",a1:"Sim. Upgrade ou downgrade num clique; cobrança proporcional e os seus funcionários IA, dados, CRM e site são preservados.",q2:"E se eu ultrapassar os tokens?",a2:"O uso extra passa a pagamento por uso com um teto mensal que você controla. Nada é cobrado acima do seu limite.",q3:"Preciso de conhecimentos técnicos?",a3:"Não. Você descreve o seu negócio e o Orquestrador Mestre contrata e configura os funcionários IA certos.",q4:"Os meus dados ficam isolados?",a4:"Sim. Cada empresa é isolada ao nÃvel da base de dados com row-level security. O Enterprise acrescenta SSO/SAML, VPC/on-prem e logs de auditoria.",q5:"Existem custos fora do plano?",a5:"Sim, e indicamo-los com clareza: nome de domÃnio, alojamento ou e-mail profissional externo, subscrições pagas dos conectores de terceiros que ligar, as suas chaves de API (BYOK), excedente de tokens, impostos locais e serviços opcionais. O orçamento final depende também das suas iterações e do consumo de tokens. Os contratos, licenças e obrigações legais necessários para que a sua empresa esteja registada, em conformidade e operacional no seu paÃs (registo comercial, seguros, fiscalidade local, RGPD, licenças profissionais, etc.) também não estão incluÃdos e ficam a seu cargo.",q6:"Quem é responsável pelo que é feito na plataforma?",a6:"O utilizador é o único responsável pelas ações, decisões e conteúdos gerados através da plataforma Infernex AI. A Infernex disponibiliza suporte técnico 24/7 por e-mail, mas declina qualquer responsabilidade por utilização indevida do serviço, por qualquer utilização não conforme com as leis aplicáveis ou com os nossos termos de utilização, bem como pelas consequências comerciais, financeiras ou legais daà resultantes."},d:{employees:"Agentes IA autónomos que você contrata, cada um com função, memória e objetivos próprios.",languages:"Idiomas em que os funcionários IA, o site e as respostas aos clientes são entregues.",website:"Um site corporativo gerado, hospedado e atualizado pela sua força de trabalho IA.",crm:"Contactos, negócios e pipeline geridos automaticamente pelos funcionários IA.",connectors:"Ferramentas de terceiros que os seus funcionários IA podem ler e acionar.",automations:"Fluxos recorrentes executados 24/7 sem supervisão.",orchestrator:"O agente mestre que distribui o trabalho, arbitra prioridades e supervisiona cada funcionário.",tokens:"Computação IA incluÃda por mês. Excedente em pagamento por uso com teto rÃgido.",seats:"Membros humanos que podem entrar e gerir a força de trabalho.",api:"Endpoint compatÃvel com OpenAI para chamar a inferência Infernex a partir do seu código.",analytics:"Horas poupadas, custo por tarefa e ROI medidos em tempo real.",security:"Isolamento de dados, controlos de acesso e garantias de conformidade.",support:"A rapidez com que um humano responde quando é importante."}}},zh:{wfpd:{seoTitle:"{{plan}} å¥é¤ â AI Workforce ä»·æ ¼ä¸åè½",seoDesc:"Infernex {{plan}} å¥é¤å®æ´è¯¦æ
ï¼AI åå·¥ãè¯è¨ãç½ç«ãCRMãè¿æ¥å¨ã令çãå®å
¨ä¸æ¯æã",eyebrow:"å¥é¤è¯¦æ
",back:"å
¨é¨å¥é¤",intro:"{{plan}} å¥é¤å
å«çä¸åï¼éæ¡è¯´æï¼æ¯«æ éèã",includedTitle:"éæ¡ååºä½ å°è·å¾çå
容",notIncludedTitle:"ä¸å
å«çå
容",notIncludedText:"æ´é«å¥é¤å¯ç¨ â éæ¶åçº§ï¼æ°æ®å®æ´ä¿çã",ctaTitle:"åå¤å¥½é¨ç½²ä½ ç AI åå·¥å¢éäºåï¼",ctaText:"å建å
¬å¸ï¼éä½£ç¬¬ä¸æ¹ AI åå·¥ï¼å åéå
å³å¯ä¸çº¿ã",ctaCompare:"ä¸å
¶ä»å¥é¤å¯¹æ¯",trust1:"30 天鿬¾ä¿è¯",trust2:"æ éä¿¡ç¨å¡",trust3:"éæ¶åæ¶",faqTitle:"常è§é®é¢",faq:{q1:"ä¹åå¯ä»¥æ´æ¢å¥é¤åï¼",a1:"å¯ä»¥ãä¸é®å级æéçº§ï¼ææ¯ä¾è®¡è´¹ï¼AI åå·¥ãæ°æ®ãCRM åç½ç«å
¨é¨ä¿çã",q2:"è¶
åºä»¤çé¢åº¦ä¼ææ ·ï¼",a2:"è¶
åºé¨åæç¨é计费ï¼å¹¶è®¾æä½ èªå·±æ§å¶çæåº¦ç¡¬æ§ä¸éï¼ç»ä¸ä¼è¶
è¿é颿£è´¹ã",q3:"éè¦ææ¯è½ååï¼",a3:"ä¸éè¦ãä½ æè¿°ä¸å¡ï¼ä¸»ç¼æä»£çä¼èªå¨é佣并é
ç½®åéç AI åå·¥ã",q4:"æçæ°æ®æ¯é离çåï¼",a4:"æ¯ãæ¯å®¶å
¬å¸å¨æ°æ®åºå±é¢éè¿è¡çº§å®å
¨é离ï¼Enterprise å¦å SSO/SAMLãVPC/æ¬å°é¨ç½²å审计æ¥å¿ã",q5:"å¥é¤ä¹å¤è¿æè´¹ç¨åï¼",a5:"æï¼
811æä»¬å¦è¯ååºï¼ååãå¤é¨æç®¡æä¼ä¸é®ç®±ãæ¨æ¥å
¥çç¬¬ä¸æ¹è¿æ¥å¨çä»è´¹è®¢é
ãæ¨èªå¸¦ç API å¯é¥ï¼BYOKï¼ãè¶
é¢ Tokenãå½å°ç¨è´¹ä»¥åå¯éæå¡ãæç»é¢ç®è¿åå³äºæ¨çè¿ä»£æ¬¡æ°å Token æ¶èãæ¤å¤ï¼ä½¿æ¨çå
¬å¸å¨æå¨å½å®¶å®ææ³¨åãåè§å¹¶æ£å¸¸è¿è¥æéçååã许å¯è¯åæ³å¾ä¹å¡ï¼å·¥å注åãä¿é©ãå½å°ç¨å¡ãGDPR åè§ãèä¸è®¸å¯çï¼åæ ·ä¸å
å«å¨å
ï¼éç±æ¨èªè¡æ¿æ
ã",q6:"å¹³å°ä¸çæä½ç±è°è´è´£ï¼",a6:"ç¨æ·å¯¹éè¿ Infernex AI å¹³å°äº§ççæææä½ãå³çåå
å®¹æ¿æ
å
¨é¨è´£ä»»ãInfernex æä¾ 7Ã24 å°æ¶çµåé®ä»¶ææ¯æ¯æï¼ä½å¯¹æå¡çä¸å½ä½¿ç¨ãä¸ç¬¦åéç¨æ³å¾ææä»¬ä½¿ç¨æ¡æ¬¾ç使ç¨è¡ä¸ºï¼ä»¥åç±æ¤äº§ççåä¸ãè´¢å¡ææ³å¾åææ¦ä¸è´è´£ã"},d:{employees:"ä½ éä½£çèªä¸» AI 代çï¼åèªæ¥æè§è²ãè®°å¿åç®æ ã",languages:"AI åå·¥ãç½ç«å客æ·åå¤æä½¿ç¨çè¯è¨ã",website:"ç± AI å¢éçæãæç®¡å¹¶æç»æ´æ°çä¼ä¸ç½ç«ã",crm:"ç± AI åå·¥èªå¨ç»´æ¤çè系人ãåæºä¸éå®ç®¡éã",connectors:"AI åå·¥å¯è¯»åå¹¶æä½çç¬¬ä¸æ¹å·¥å
·ã",automations:"åå·¥å
¨å¤©åèªå¨è¿è¡çéå¤å·¥ä½æµã",orchestrator:"åé
ä»»å¡ã仲è£ä¼å
级并çç£æ¯ä½åå·¥ç主代çã",tokens:"æ¯æå
å«ç AI ç®åï¼è¶
åºé¨åæç¨é计费并设硬æ§ä¸éã",seats:"å¯ç»å½å¹¶ç®¡ç AI å¢éç人类æåã",api:"å
¼å®¹ OpenAI çæ¥å£ï¼å¯å¨ä½ èªå·±ç代ç ä¸è°ç¨ Infernex æ¨çã",analytics:"宿¶è¡¡éèççå·¥æ¶ãå任塿æ¬ä¸ ROIã",security:"æ°æ®é离ãè®¿é®æ§å¶ä¸åè§ä¿éã",support:"å
³é®æ¶å»è·å¾äººå·¥ååºçé度ã"}}},ja:{wfpd:{seoTitle:"{{plan}} ãã©ã³ â AI Workforce ã®æéã¨æ©è½",seoDesc:"Infernex {{plan}} ãã©ã³ã®è©³ç´°ï¼AI 徿¥å¡ãè¨èªãã¦ã§ããµã¤ããCRMãã³ãã¯ã¿ããã¼ã¯ã³ãã»ãã¥ãªãã£ããµãã¼ãã",eyebrow:"ãã©ã³è©³ç´°",back:"ãã¹ã¦ã®ãã©ã³",intro:"{{plan}} ã§å¾ããããã®ãä¸é
ç®ãã¤æè¨ãã¾ããæ³å®å¤ã¯ããã¾ããã",includedTitle:"å¾ãããå
容ï¼é
ç®å¥ï¼",notIncludedTitle:"å«ã¾ããªããã®",notIncludedText:"ä¸ä½ãã©ã³ã§å©ç¨å¯è½ â ãã¤ã§ãã¢ããã°ã¬ã¼ãã§ãããã¼ã¿ã¯ãã®ã¾ã¾å¼ãç¶ããã¾ãã",ctaTitle:"AI 徿¥å¡ãå°å
¥ããæºåã¯ã§ãã¾ãããï¼",ctaText:"ä¼ç¤¾ã使ããæåã® AI 徿¥å¡ãæ¡ç¨ãã¦æ°åã§ç¨¼åéå§ã",ctaCompare:"ä»ã®ãã©ã³ã¨æ¯è¼",trust1:"30æ¥éè¿éä¿è¨¼",trust2:"ã¯ã¬ã¸ããã«ã¼ãä¸è¦",trust3:"ãã¤ã§ãè§£ç´å¯è½",faqTitle:"ãããã質å",faq:{q1:"å¾ãããã©ã³å¤æ´ã§ãã¾ããï¼",a1:"ã¯ããã¯ã³ã¯ãªãã¯ã§ä¸ä½ã»ä¸ä½ã«å¤æ´ã§ããæ¥å²ãè«æ±ã§ AI 徿¥å¡ã»ãã¼ã¿ã»CRMã»ãµã¤ãã¯ãã¹ã¦ä¿æããã¾ãã",q2:"ãã¼ã¯ã³ãè¶
éãããï¼",a2:"è¶
éåã¯å¾é課éã¨ãªãããèªèº«ã§è¨å®ããæéä¸éãè¶
ãã¦è«æ±ããããã¨ã¯ããã¾ããã",q3:"æè¡ç¥èã¯å¿
è¦ã§ããï¼",a3:"ä¸è¦ã§ããäºæ¥å
容ã説æããã°ããã¹ã¿ã¼ãªã¼ã±ã¹ãã¬ã¼ã¿ã¼ãé©å㪠AI 徿¥å¡ãæ¡ç¨ã»è¨å®ãã¾ãã",q4:"ãã¼ã¿ã¯åé¢ããã¾ããï¼",a4:"ã¯ããå社ã¯ãã¼ã¿ãã¼ã¹ã¬ãã«ã®è¡ã¬ãã«ã»ãã¥ãªãã£ã§åé¢ããã¾ããEnterprise ã§ã¯ SSO/SAMLãVPC/ãªã³ãã¬ãç£æ»ãã°ã追å ãã¾ãã",q5:"ãã©ã³å¤ã®è²»ç¨ã¯ããã¾ããï¼",a5:"ã¯ããæç¤ºãã¦ãã¾ãï¼ãã¡ã¤ã³åãå¤é¨ãã¹ãã£ã³ã°ãæ³äººã¡ã¼ã«ãæ¥ç¶ãã第ä¸è
ã³ãã¯ã¿ã®ææãµãã¹ã¯ãªãã·ã§ã³ãæã¡è¾¼ã¿ API ãã¼ï¼BYOKï¼ããã¼ã¯ã³è¶
éåãç¾å°ç¨ããªãã·ã§ã³ãµã¼ãã¹ã§ããæçµäºç®ã¯å復忰ã¨ãã¼ã¯ã³æ¶è²»éã«ãå·¦å³ããã¾ããã¾ãã貴社ãèªå½ã§ç»è¨ãããæ³ä»¤ãéµå®ããäºæ¥ãéå¶ããããã«å¿
è¦ãªå¥ç´ã»ã©ã¤ã»ã³ã¹ã»æ³ç義åï¼åæ¥ç»è¨ãä¿éºãç¾å°ç¨åãGDPR対å¿ãå°éã©ã¤ã»ã³ã¹çï¼ãå«ã¾ãããã客æ§ã®ãè² æ
ã¨ãªãã¾ãã",q6:"ãã©ãããã©ã¼ã ä¸ã®è¡çºã®è²¬ä»»ã¯èª°ã«ããã¾ããï¼",a6:"Infernex AI ãã©ãããã©ã¼ã ãéãã¦çæãããè¡çºã»å¤æã»ã³ã³ãã³ãã«ã¤ãã¦ã¯ãã¦ã¼ã¶ã¼ãåç¬ã§è²¬ä»»ãè² ãã¾ããInfernex ã¯ã¡ã¼ã«ã«ãã 24 æé 365 æ¥ã®æè¡ãµãã¼ããæä¾ãã¾ããããµã¼ãã¹ã®ä¸é©åã
811ªå©ç¨ãé©ç¨æ³ä»¤ã¾ãã¯å©ç¨è¦ç´ã«åããå©ç¨ãããã³ãããã«èµ·å ãã忥çã»è²¡åçã»æ³ççµæã«ã¤ãã¦ä¸åã®è²¬ä»»ãè² ãã¾ããã"},d:{employees:"å½¹å²ã»è¨æ¶ã»ç®æ¨ãæã¤èªå¾å AI ã¨ã¼ã¸ã§ã³ããæ¡ç¨ã§ãã¾ãã",languages:"AI 徿¥å¡ã»ã¦ã§ããµã¤ãã»é¡§å®¢å¯¾å¿ãæä¾ãããè¨èªã§ãã",website:"AI 徿¥å¡ãçæã»ãã¹ãã£ã³ã°ã»æ´æ°ãç¶ãã伿¥ãµã¤ãã",crm:"é£çµ¡å
ã»åè«ã»ãã¤ãã©ã¤ã³ã AI 徿¥å¡ãèªåã§ç®¡çãã¾ãã",connectors:"AI 徿¥å¡ãåç
§ã»æä½ã§ããå¤é¨ãã¼ã«ã",automations:"ç£è¦ãªãã§ 24/7 å®è¡ãããå®åã¯ã¼ã¯ããã¼ã",orchestrator:"æ¥åãå²ãå½ã¦ãåªå
度ãè£å®ããå
¨å¾æ¥å¡ãç£ç£ãããã¹ã¿ã¼ã¨ã¼ã¸ã§ã³ãã",tokens:"æ¯æå«ã¾ãã AI è¨ç®éãè¶
éåã¯ä¸éä»ãã®å¾é課éã",seats:"ãã°ã¤ã³ã㦠AI çµç¹ãéç¨ã§ãã人éã®ã¡ã³ãã¼æ°ã",api:"èªç¤¾ã³ã¼ããã Infernex æ¨è«ãå¼ã¹ã OpenAI äºæã¨ã³ããã¤ã³ãã",analytics:"忏æéã»ã¿ã¹ã¯å価ã»ROI ããªã¢ã«ã¿ã¤ã ã§è¨æ¸¬ã",security:"ãã¼ã¿åé¢ãã¢ã¯ã»ã¹å¶å¾¡ãã³ã³ãã©ã¤ã¢ã³ã¹ä¿è¨¼ã",support:"éè¦ãªã¨ãã«äººéã®åçãå¾ãããéãã"}}},hi:{wfpd:{seoTitle:"{{plan}} पà¥à¤²à¤¾à¤¨ â AI Workforce मà¥à¤²à¥à¤¯ à¤à¤° सà¥à¤µà¤¿à¤§à¤¾à¤à¤",seoDesc:"Infernex {{plan}} पà¥à¤²à¤¾à¤¨ à¤à¤¾ पà¥à¤°à¤¾ विवरण: AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥, à¤à¤¾à¤·à¤¾à¤à¤, वà¥à¤¬à¤¸à¤¾à¤à¤, CRM, à¤à¤¨à¥à¤à¥à¤à¤°, à¤à¥à¤à¤¨, सà¥à¤°à¤à¥à¤·à¤¾ à¤à¤° सहायता।",eyebrow:"पà¥à¤²à¤¾à¤¨ विवरण",back:"सà¤à¥ पà¥à¤²à¤¾à¤¨",intro:"{{plan}} à¤à¥ साथ à¤à¤ªà¤à¥ à¤à¥à¤¯à¤¾-à¤à¥à¤¯à¤¾ मिलता हॠâ बिà¤à¤¦à¥à¤µà¤¾à¤°, बिना à¤à¤¿à¤¸à¥ à¤à¤¿à¤ªà¥ शरà¥à¤¤ à¤à¥à¥¤",includedTitle:"à¤à¤ªà¤à¥ à¤à¥à¤¯à¤¾ मिलता हà¥, बिà¤à¤¦à¥ दर बिà¤à¤¦à¥",notIncludedTitle:"à¤à¥ शामिल नहà¥à¤ हà¥",notIncludedText:"à¤à¤à¥à¤ पà¥à¤²à¤¾à¤¨ मà¥à¤ à¤à¤ªà¤²à¤¬à¥à¤§ â à¤à¤à¥ à¤à¥ à¤
पà¤à¥à¤°à¥à¤¡ à¤à¤°à¥à¤, à¤à¤ªà¤à¤¾ डà¥à¤à¤¾ साथ à¤à¤²à¤¤à¤¾ हà¥à¥¤",ctaTitle:"à¤
पनॠAI वरà¥à¤à¤«à¥à¤°à¥à¤¸ शà¥à¤°à¥ à¤à¤°à¤¨à¥ à¤à¥ लिठतà¥à¤¯à¤¾à¤° हà¥à¤?",ctaText:"à¤
पनॠà¤à¤à¤ªà¤¨à¥ बनाà¤à¤, पहलॠAI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ नियà¥à¤à¥à¤¤ à¤à¤°à¥à¤ à¤à¤° मिनà¤à¥à¤ मà¥à¤ लाà¤à¤µ à¤à¤¾à¤à¤à¥¤",ctaCompare:"à¤
नà¥à¤¯ पà¥à¤²à¤¾à¤¨ सॠतà¥à¤²à¤¨à¤¾ à¤à¤°à¥à¤",trust1:"30-दिन मनà¥-बà¥à¤ à¤à¤¾à¤°à¤à¤à¥",trust2:"à¤à¥à¤°à¥à¤¡à¤¿à¤ à¤à¤¾à¤°à¥à¤¡ à¤à¥ à¤à¤¼à¤°à¥à¤°à¤¤ नहà¥à¤",trust3:"à¤à¤à¥ à¤à¥ रदà¥à¤¦ à¤à¤°à¥à¤",faqTitle:"à¤
à¤à¥à¤¸à¤° पà¥à¤à¥ à¤à¤¾à¤¨à¥ वालॠपà¥à¤°à¤¶à¥à¤¨",faq:{q1:"à¤à¥à¤¯à¤¾ मà¥à¤ बाद मà¥à¤ पà¥à¤²à¤¾à¤¨ बदल सà¤à¤¤à¤¾ हà¥à¤?",a1:"हाà¤à¥¤ à¤à¤ à¤à¥à¤²à¤¿à¤ मà¥à¤ à¤
पà¤à¥à¤°à¥à¤¡ या डाà¤à¤¨à¤à¥à¤°à¥à¤¡; बिलिà¤à¤ à¤à¤¨à¥à¤ªà¤¾à¤¤à¤¿à¤ हà¥à¤¤à¥ हॠà¤à¤° à¤à¤ªà¤à¥ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥, डà¥à¤à¤¾, CRM व वà¥à¤¬à¤¸à¤¾à¤à¤ सà¥à¤°à¤à¥à¤·à¤¿à¤¤ रहतॠहà¥à¤à¥¤",q2:"à¤à¥à¤à¤¨ सà¥à¤®à¤¾ पार हà¥à¤¨à¥ पर à¤à¥à¤¯à¤¾ हà¥à¤à¤¾?",a2:"à¤
तिरिà¤à¥à¤¤ à¤à¤ªà¤¯à¥à¤ पà¥-à¤à¤à¤¼-यà¥-à¤à¥ पर à¤à¤²à¤¾ à¤à¤¾à¤¤à¤¾ हà¥, मासिठहारà¥à¤¡ à¤à¥à¤ª à¤à¥ साथ à¤à¤¿à¤¸à¥ à¤à¤ª नियà¤à¤¤à¥à¤°à¤¿à¤¤ à¤à¤°à¤¤à¥ हà¥à¤à¥¤ सà¥à¤®à¤¾ सॠà¤à¤à¥ à¤à¥à¤ à¤à¥ बिल नहà¥à¤ हà¥à¤¤à¤¾à¥¤",q3:"à¤à¥à¤¯à¤¾ तà¤à¤¨à¥à¤à¥ à¤à¥à¤¶à¤² à¤à¤¾à¤¹à¤¿à¤?",a3:"नहà¥à¤à¥¤ à¤à¤ª à¤
पना वà¥à¤¯à¤µà¤¸à¤¾à¤¯ बताà¤à¤, मासà¥à¤à¤° à¤à¤°à¥à¤à¥à¤¸à¥à¤à¥à¤°à¥à¤à¤° सहॠAI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ नियà¥à¤à¥à¤¤ à¤à¤° à¤à¥à¤¨à¥à¤«à¤¼à¤¿à¤à¤° à¤à¤° दà¥à¤¤à¤¾ हà¥à¥¤",q4:"à¤à¥à¤¯à¤¾ मà¥à¤°à¤¾ डà¥à¤à¤¾ à¤
लठरहता हà¥?",a4:"हाà¤à¥¤ हर à¤à¤à¤ªà¤¨à¥ डà¥à¤à¤¾à¤¬à¥à¤¸ सà¥à¤¤à¤° पर row-level security सॠà¤
लठरहतॠहà¥à¥¤ Enterprise मà¥à¤ SSO/SAML, VPC/on-prem à¤à¤° à¤à¤¡à¤¿à¤ लà¥à¤ à¤à¥ मिलतॠहà¥à¤à¥¤",q5:"à¤à¥à¤¯à¤¾ पà¥à¤²à¤¾à¤¨ à¤à¥ बाहर à¤à¥ à¤à¤°à¥à¤ हà¥à¤?",a5:"हाà¤, à¤à¤° हम à¤à¤¨à¥à¤¹à¥à¤ सà¥à¤ªà¤·à¥à¤ बतातॠहà¥à¤: डà¥à¤®à¥à¤¨ नाम, बाहरॠहà¥à¤¸à¥à¤à¤¿à¤à¤ या पà¥à¤°à¥à¤«à¥à¤¶à¤¨à¤² à¤à¤®à¥à¤², à¤à¤ªà¤à¥ à¤à¥à¤¡à¤¼à¥ à¤à¤ तà¥à¤¤à¥à¤¯-पà¤à¥à¤· à¤à¤¨à¥à¤à¥à¤à¤°à¥à¤ à¤à¥ सशà¥à¤²à¥à¤ सदसà¥à¤¯à¤¤à¤¾, à¤à¤ªà¤à¥ à¤
पनॠAPI à¤à¥à¤à¤à¤¿à¤¯à¤¾à¤ (BYOK), à¤à¥à¤à¤¨ à¤
धिशà¥à¤·, सà¥à¤¥à¤¾à¤¨à¥à¤¯ à¤à¤° à¤à¤° वà¥à¤à¤²à¥à¤ªà¤¿à¤ सà¥à¤µà¤¾à¤à¤à¥¤ à¤
à¤à¤¤à¤¿à¤® बà¤à¤ à¤à¤ªà¤à¥ पà¥à¤¨à¤°à¤¾à
811¤µà¥à¤¤à¥à¤¤à¤¿à¤¯à¥à¤ à¤à¤° à¤à¥à¤à¤¨ à¤à¤ªà¤¤ पर à¤à¥ निरà¥à¤à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤ à¤à¤¸à¤à¥ à¤
लावा, à¤à¤ªà¤à¥ à¤à¤à¤ªà¤¨à¥ à¤à¥ à¤à¤ªà¤à¥ दà¥à¤¶ मà¥à¤ पà¤à¤à¥à¤à¥à¤¤, à¤
नà¥à¤ªà¤¾à¤²à¤¨à¤¶à¥à¤² à¤à¤° सà¤à¤à¤¾à¤²à¤¨ यà¥à¤à¥à¤¯ बनानॠà¤à¥ लिठà¤à¤µà¤¶à¥à¤¯à¤ à¤
नà¥à¤¬à¤à¤§, लाà¤à¤¸à¥à¤à¤¸ à¤à¤° à¤à¤¾à¤¨à¥à¤¨à¥ दायितà¥à¤µ (à¤à¤à¤ªà¤¨à¥ पà¤à¤à¥à¤à¤°à¤£, बà¥à¤®à¤¾, सà¥à¤¥à¤¾à¤¨à¥à¤¯ à¤à¤°, GDPR à¤
नà¥à¤ªà¤¾à¤²à¤¨, पà¥à¤¶à¥à¤µà¤° लाà¤à¤¸à¥à¤à¤¸ à¤à¤¦à¤¿) à¤à¥ शामिल नहà¥à¤ हà¥à¤ à¤à¤° à¤à¤ªà¤à¥ à¤à¤°à¥à¤ पर रहतॠहà¥à¤à¥¤",q6:"पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤® पर à¤à¤¿à¤ à¤à¤ à¤à¤¾à¤°à¥à¤¯à¥à¤ à¤à¥ लिठà¤à¥à¤¨ à¤à¤¤à¥à¤¤à¤°à¤¦à¤¾à¤¯à¥ हà¥?",a6:"Infernex AI पà¥à¤²à¥à¤à¤«à¤¼à¥à¤°à¥à¤® à¤à¥ माधà¥à¤¯à¤® सॠà¤à¤¿à¤ à¤à¤ à¤à¤¾à¤°à¥à¤¯à¥à¤, निरà¥à¤£à¤¯à¥à¤ à¤à¤° à¤à¤¤à¥à¤ªà¤¨à¥à¤¨ सामà¤à¥à¤°à¥ à¤à¥ लिठपà¥à¤°à¥à¤£ रà¥à¤ª सॠà¤à¤ªà¤¯à¥à¤à¤à¤°à¥à¤¤à¤¾ हॠà¤à¤¤à¥à¤¤à¤°à¤¦à¤¾à¤¯à¥ हà¥à¥¤ Infernex à¤à¤®à¥à¤² दà¥à¤µà¤¾à¤°à¤¾ 24/7 तà¤à¤¨à¥à¤à¥ सहायता पà¥à¤°à¤¦à¤¾à¤¨ à¤à¤°à¤¤à¤¾ हà¥, परà¤à¤¤à¥ सà¥à¤µà¤¾ à¤à¥ दà¥à¤°à¥à¤ªà¤¯à¥à¤, लाà¤à¥ à¤à¤¾à¤¨à¥à¤¨à¥à¤ या हमारॠà¤à¤ªà¤¯à¥à¤ शरà¥à¤¤à¥à¤ à¤à¥ विरà¥à¤¦à¥à¤§ à¤à¤ªà¤¯à¥à¤, तथा à¤à¤¸à¤¸à¥ à¤à¤¤à¥à¤ªà¤¨à¥à¤¨ वà¥à¤¯à¤¾à¤µà¤¸à¤¾à¤¯à¤¿à¤, वितà¥à¤¤à¥à¤¯ या à¤à¤¾à¤¨à¥à¤¨à¥ परिणामà¥à¤ à¤à¥ लिठà¤à¤¿à¤¸à¥ à¤à¥ दायितà¥à¤µ सॠà¤à¤¨à¤à¤¾à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤"},d:{employees:"सà¥à¤µà¤¾à¤¯à¤¤à¥à¤¤ AI à¤à¤à¥à¤à¤ à¤à¤¿à¤¨à¥à¤¹à¥à¤ à¤à¤ª नियà¥à¤à¥à¤¤ à¤à¤°à¤¤à¥ हà¥à¤ â à¤
पनॠà¤à¥à¤®à¤¿à¤à¤¾, सà¥à¤®à¥à¤¤à¤¿ à¤à¤° लà¤à¥à¤·à¥à¤¯à¥à¤ à¤à¥ साथ।",languages:"à¤à¤¿à¤¨ à¤à¤¾à¤·à¤¾à¤à¤ मà¥à¤ à¤à¤ªà¤à¥ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥, वà¥à¤¬à¤¸à¤¾à¤à¤ à¤à¤° à¤à¥à¤°à¤¾à¤¹à¤ à¤à¤¤à¥à¤¤à¤° à¤à¤ªà¤²à¤¬à¥à¤§ हà¥à¤¤à¥ हà¥à¤à¥¤",website:"à¤à¤ªà¤à¥ AI वरà¥à¤à¤«à¥à¤°à¥à¤¸ दà¥à¤µà¤¾à¤°à¤¾ बनाà¤, हà¥à¤¸à¥à¤ à¤à¤° à¤
पडà¥à¤ à¤à¥ à¤à¤¾à¤¨à¥ वालॠà¤à¤à¤ªà¤¨à¥ वà¥à¤¬à¤¸à¤¾à¤à¤à¥¤",crm:"सà¤à¤ªà¤°à¥à¤, सà¥à¤¦à¥ à¤à¤° पाà¤à¤ªà¤²à¤¾à¤à¤¨ à¤à¤ªà¤à¥ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ दà¥à¤µà¤¾à¤°à¤¾ सà¥à¤µà¤¤à¤ सà¤à¤à¤¾à¤²à¥ à¤à¤¾à¤¤à¥ हà¥à¤à¥¤",connectors:"तà¥à¤¸à¤°à¥ पà¤à¥à¤· à¤à¥ à¤à¥à¤² à¤à¤¿à¤¨à¥à¤¹à¥à¤ à¤à¤ªà¤à¥ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ पढ़ à¤à¤° सà¤à¤à¤¾à¤²à¤¿à¤¤ à¤à¤° सà¤à¤¤à¥ हà¥à¤à¥¤",automations:"नियमित वरà¥à¤à¤«à¤¼à¥à¤²à¥ à¤à¥ बिना निà¤à¤°à¤¾à¤¨à¥ 24/7 à¤à¤²à¤¤à¥ हà¥à¤à¥¤",orchestrator:"मासà¥à¤à¤° à¤à¤à¥à¤à¤ à¤à¥ à¤à¤¾à¤°à¥à¤¯ बाà¤à¤à¤¤à¤¾, पà¥à¤°à¤¾à¤¥à¤®à¤¿à¤à¤¤à¤¾à¤à¤ तय à¤à¤°à¤¤à¤¾ à¤à¤° हर à¤à¤°à¥à¤®à¤à¤¾à¤°à¥ à¤à¥ निà¤à¤°à¤¾à¤¨à¥ à¤à¤°à¤¤à¤¾ हà¥à¥¤",tokens:"हर माह शामिल AI à¤à¤à¤ªà¥à¤¯à¥à¤à¥¤ à¤
तिरिà¤à¥à¤¤ à¤à¤ªà¤¯à¥à¤ हारà¥à¤¡ à¤à¥à¤ª à¤à¥ साथ पà¥-à¤à¤à¤¼-यà¥-à¤à¥à¥¤",seats:"मानव सदसà¥à¤¯ à¤à¥ लà¥à¤à¤¿à¤¨ à¤à¤° वरà¥à¤à¤«à¥à¤°à¥à¤¸ à¤à¤¾ सà¤à¤à¤¾à¤²à¤¨ à¤à¤° सà¤à¤¤à¥ हà¥à¤à¥¤",api:"OpenAI-सà¤à¤à¤¤ à¤à¤à¤¡à¤ªà¥à¤à¤à¤, à¤
पनॠà¤à¥à¤¡ सॠInfernex à¤à¤¨à¥à¤«à¤°à¥à¤à¤¸ à¤à¥à¤² à¤à¤°à¥à¤à¥¤",analytics:"बà¤à¤¾à¤ à¤à¤ à¤à¤à¤à¥, पà¥à¤°à¤¤à¤¿ à¤à¤¾à¤°à¥à¤¯ लाà¤à¤¤ à¤à¤° ROI वासà¥à¤¤à¤µà¤¿à¤ समय मà¥à¤à¥¤",security:"डà¥à¤à¤¾ पà¥à¤¥à¤à¥à¤à¤°à¤£, à¤à¤à¥à¤¸à¥à¤¸ नियà¤à¤¤à¥à¤°à¤£ à¤à¤° à¤
नà¥à¤ªà¤¾à¤²à¤¨ à¤à¤¾à¤°à¤à¤à¥à¥¤",support:"à¤à¤¼à¤°à¥à¤°à¤¤ à¤à¥ समय मानव à¤à¤¤à¥à¤¤à¤° à¤à¤¿à¤¤à¤¨à¥ à¤à¤²à¥à¤¦à¥ मिलता हà¥à¥¤"}}},ar:{wfpd:{seoTitle:"خطة {{plan}} â Ø£Ø³Ø¹Ø§Ø± ÙÙ
Ø²Ø§ÙØ§ AI Workforce",seoDesc:"ØªÙØ§ØµÙÙ ÙØ§Ù
ÙØ© ÙØ®Ø·Ø© Infernex {{plan}}: Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹ÙØ اÙÙØºØ§ØªØ اÙÙ
ÙÙØ¹Ø CRMØ Ø§ÙÙ
ÙØµÙØ§ØªØ Ø§ÙØ±Ù
ÙØ²Ø Ø§ÙØ£Ù
Ø§Ù ÙØ§ÙدعÙ
.",eyebrow:"ØªÙØ§ØµÙÙ Ø§ÙØ®Ø·Ø©",back:"ÙÙ Ø§ÙØ®Ø·Ø·",intro:"ÙÙ Ù
ا ØªØØµÙ عÙÙÙ Ù
ع {{plan}}Ø ÙÙØ·Ø© بÙÙØ·Ø© â Ø¨Ùا Ù
ÙØ§Ø¬Ø¢Øª.",includedTitle:"Ù
ا ØªØØµÙ عÙÙÙØ ÙÙØ·Ø© بÙÙØ·Ø©",notIncludedTitle:"Ù
ا ÙÙ ØºÙØ± Ù
Ø´Ù
ÙÙ",notIncludedText:"Ù
ØªØ§Ø Ù٠خطة أعÙÙ â ÙÙ
ÙÙÙ Ø§ÙØªØ±ÙÙØ© Ù٠أ٠ÙÙØª ÙØªÙتÙÙ Ø¨ÙØ§Ùات٠Ù
عÙ.",ctaTitle:"Ø¬Ø§ÙØ² ÙÙØ´Ø± ÙÙØ© Ø§ÙØ¹Ù
Ù Ø§ÙØ°ÙÙØ© Ø§ÙØ®Ø§ØµØ© Ø¨ÙØ",ctaText:"Ø£ÙØ´Ø¦ Ø´Ø±ÙØªÙØ ÙØ¸Ù٠أÙÙ Ù
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù ÙØ§ÙØ·ÙÙ Ø®ÙØ§Ù Ø¯ÙØ§Ø¦Ù.",ctaCompare:"ÙØ§Ø±Ù Ù
ع Ø§ÙØ®Ø·Ø· Ø§ÙØ£Ø®Ø±Ù",trust1:"ضÙ
ا٠استرداد اÙÙ
Ø§Ù Ø®ÙØ§Ù 30 ÙÙÙ
ÙØ§",trust2:"بدÙÙ Ø¨Ø·Ø§ÙØ© ائتÙ
اÙ",trust3:"Ø¥ÙØºØ§Ø¡ Ù٠أ٠ÙÙØª",faqTitle:"Ø§ÙØ£Ø³Ø¦ÙØ© Ø§ÙØ´Ø§Ø¦Ø¹Ø©",faq:{q1:"ÙÙ ÙÙ
ÙÙÙ٠تغÙÙØ± Ø§ÙØ®Ø·Ø© ÙØ§ØÙÙØ§Ø",a1:"ÙØ¹Ù
. ترÙÙØ© أ٠تخÙÙØ¶ بÙÙØ±Ø© ÙØ§ØØ¯Ø©Ø اÙÙÙØªØ±Ø© ØªÙØ§Ø³Ø¨ÙØ© ÙÙØªÙ
Ø§ÙØ§ØØªÙاظ بÙ
ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù ÙØ§ÙØ¨ÙØ§Ùات ÙCRM ÙØ§ÙÙ
ÙÙØ¹.",q2:"Ù
اذا ÙÙ ØªØ¬Ø§ÙØ²Øª Ø§ÙØ±Ù
ÙØ²Ø",a2:"ÙØªØÙÙ Ø§ÙØ§Ø³ØªÙÙØ§Ù Ø§ÙØ¥Ø¶Ø§Ù٠إÙÙ Ø§ÙØ¯Ùع ØØ³Ø¨ Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù
Ù
ع ØØ¯ Ø´ÙØ±Ù صارÙ
تتØÙÙ
ÙÙÙ. ÙØ§ Ø´ÙØ¡ ÙÙÙÙØªØ± ÙÙÙ ØØ¯ÙÙ.",q3:"ÙÙ Ø£ØØªØ§Ø¬ Ù
ÙØ§Ø±Ø§Øª تÙÙÙØ©Ø",a3:"ÙØ§. ØªØµÙ ÙØ´Ø§Ø·Ù ÙÙÙÙ
811٠اÙÙ ÙØ³ÙÙ Ø§ÙØ±Ø¦ÙØ³Ù Ø¨ØªÙØ¸ÙÙ ÙØªÙÙØ¦Ø© Ù ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·Ùاع٠اÙÙ ÙØ§Ø³Ø¨ÙÙ.",q4:"ÙÙ Ø¨ÙØ§Ùات٠٠عزÙÙØ©Ø",a4:"ÙØ¹Ù . ÙÙ Ø´Ø±ÙØ© ٠عزÙÙØ© عÙ٠٠ستÙÙ ÙØ§Ø¹Ø¯Ø© Ø§ÙØ¨ÙØ§ÙØ§Øª عبر row-level security. ÙØªØ¶Ù٠خطة Enterprise تسجÙÙ Ø§ÙØ¯Ø®Ù٠اÙÙ ÙØØ¯ ÙVPC/on-prem ÙØ³Ø¬Ùات Ø§ÙØªØ¯ÙÙÙ.",q5:"ÙÙ ØªÙØ¬Ø¯ ØªÙØ§ÙÙ٠خارج Ø§ÙØ®Ø·Ø©Ø",a5:"ÙØ¹Ù Ø ÙÙØ°ÙØ±ÙØ§ Ø¨ÙØ¶ÙØ: اس٠اÙÙØ·Ø§ÙØ Ø§ÙØ§Ø³ØªØ¶Ø§ÙØ© Ø§ÙØ®Ø§Ø±Ø¬ÙØ© Ø£Ù Ø§ÙØ¨Ø±Ùد اÙÙ ÙÙÙØ Ø§Ø´ØªØ±Ø§ÙØ§Øª اÙÙ ÙØµÙÙØ§Øª Ø§ÙØ®Ø§Ø±Ø¬ÙØ© اÙ٠دÙÙØ¹Ø© Ø§ÙØªÙ ØªØ±Ø¨Ø·ÙØ§Ø Ù ÙØ§ØªÙØ Ø§ÙÙØ§Ø¬Ùات Ø§ÙØ®Ø§ØµØ© ب٠(BYOK)Ø ØªØ¬Ø§ÙØ² Ø§ÙØ±Ù ÙØ²Ø Ø§ÙØ¶Ø±Ø§Ø¦Ø¨ اÙÙ ØÙÙØ© ÙØ§Ùخد٠ات Ø§ÙØ§Ø®ØªÙØ§Ø±ÙØ©. Ù٠ا تعت٠د اÙÙ ÙØ²Ø§ÙÙØ© اÙÙÙØ§Ø¦ÙØ© عÙ٠عدد Ø§ÙØªÙرارات ÙØ§Ø³ØªÙÙØ§Ù Ø§ÙØ±Ù ÙØ². Ù٠ا Ø£Ù Ø§ÙØ¹ÙÙØ¯ ÙØ§ÙØªØ±Ø§Ø®ÙØµ ÙØ§ÙØ§ÙØªØ²Ø§Ù ات اÙÙØ§ÙÙÙÙØ© اÙÙØ§Ø²Ù Ø© ÙØªØ³Ø¬ÙÙ Ø´Ø±ÙØªÙ ÙØ§Ù تثاÙÙØ§ ÙØªØ´ØºÙÙÙØ§ ÙÙ Ø¨ÙØ¯Ù (Ø§ÙØ³Ø¬Ù Ø§ÙØªØ¬Ø§Ø±ÙØ Ø§ÙØªØ£Ù ÙÙØ§ØªØ Ø§ÙØ¶Ø±Ø§Ø¦Ø¨ اÙÙ ØÙÙØ©Ø Ø§ÙØ§Ù تثا٠ÙÙØ§Ø¦ØØ© Ø§ÙØ¹Ø§Ù Ø© ÙØÙ Ø§ÙØ© Ø§ÙØ¨ÙØ§ÙØ§ØªØ Ø§ÙØªØ±Ø§Ø®Ùص اÙÙ ÙÙÙØ©Ø Ø¥ÙØ®) ØºÙØ± Ù Ø´Ù ÙÙØ© Ø£ÙØ¶Ø§Ù ÙØªØ¨Ù٠عÙÙ ÙÙÙØªÙ.",q6:"٠٠اÙ٠سؤÙÙ Ø¹Ù ÙØ§ ÙØªÙ عÙ٠اÙÙ ÙØµØ©Ø",a6:"ÙØªØÙ Ù٠اÙÙ Ø³ØªØ®Ø¯Ù ÙØØ¯Ù Ø§Ù٠سؤÙÙÙØ© Ø¹Ù Ø§ÙØ¥Ø¬Ø±Ø§Ø¡Ø§Øª ÙØ§ÙÙØ±Ø§Ø±Ø§Øª ÙØ§ÙÙ ØØªÙ٠اÙÙØ§ØªØ¬ عبر Ù ÙØµØ© Infernex AI. تÙÙÙØ± Infernex دع٠ا٠تÙÙÙØ§Ù عÙ٠٠دار Ø§ÙØ³Ø§Ø¹Ø© Ø·ÙØ§Ù Ø£ÙØ§Ù Ø§ÙØ£Ø³Ø¨Ùع عبر Ø§ÙØ¨Ø±Ùد Ø§ÙØ¥ÙÙØªØ±ÙÙÙØ ÙÙÙÙØ§ ØªÙØ®Ù٠٠سؤÙÙÙØªÙا Ø¹Ù Ø£Ù Ø³ÙØ¡ استخدا٠ÙÙØ®Ø¯Ù Ø© Ø£Ù Ø§Ø³ØªØ®Ø¯Ø§Ù ÙØ§ ÙØªÙاÙ٠٠ع اÙÙÙØ§ÙÙ٠اÙ٠ع٠ÙÙ Ø¨ÙØ§ Ø£Ù Ø´Ø±ÙØ· Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù Ø ÙØ¹Ù Ø£Ù Ø¹ÙØ§Ùب ØªØ¬Ø§Ø±ÙØ© أ٠٠اÙÙØ© Ø£Ù ÙØ§ÙÙÙÙØ© ØªÙØ¬Ù ع٠ذÙÙ."},d:{employees:"ÙÙÙØ§Ø¡ Ø°ÙØ§Ø¡ Ø§ØµØ·ÙØ§Ø¹Ù ٠ستÙÙÙÙ ØªÙØ¸ÙÙÙÙ Ø ÙÙÙ Ù ÙÙÙ Ø¯ÙØ±Ù ÙØ°Ø§ÙØ±ØªÙ ÙØ£ÙداÙÙ.",languages:"اÙÙØºØ§Øª Ø§ÙØªÙ ÙÙÙØ¯ÙÙÙ Ø¨ÙØ§ Ù ÙØ¸ÙÙÙ ÙÙ ÙÙØ¹Ù ÙØ±Ø¯Ùد٠عÙÙ Ø§ÙØ¹Ù ÙØ§Ø¡.",website:"Ù ÙÙØ¹ Ø´Ø±ÙØ© ÙÙÙØ´Ø£ ÙÙÙØ³ØªØ¶Ø§Ù ÙÙÙØØ¯ÙÙØ« باست٠رار Ø¨ÙØ§Ø³Ø·Ø© ÙÙØ© ع٠ÙÙ Ø§ÙØ°ÙÙØ©.",crm:"Ø¬ÙØ§Øª Ø§ÙØ§ØªØµØ§Ù ÙØ§ÙصÙÙØ§Øª ÙØ®Ø· اÙÙ Ø¨ÙØ¹Ø§Øª ÙØ¯ÙØ±ÙØ§ Ù ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·Ùاع٠تÙÙØ§Ø¦ÙÙØ§.",connectors:"Ø£Ø¯ÙØ§Øª Ø®Ø§Ø±Ø¬ÙØ© ÙÙ ÙÙ ÙÙ ÙØ¸ÙÙÙ ÙØ±Ø§Ø¡ØªÙا ÙØ§ÙتØÙÙ Ø¨ÙØ§.",automations:"Ø³ÙØ± Ø¹Ù Ù Ù ØªÙØ±Ø± ÙØ¹Ù Ù 24/7 دÙ٠إشراÙ.",orchestrator:"اÙÙÙÙÙ Ø§ÙØ±Ø¦ÙØ³Ù Ø§ÙØ°Ù ÙÙØ²Ùع اÙÙ ÙØ§Ù ÙÙØØ¯ÙØ¯ Ø§ÙØ£ÙÙÙÙØ§Øª ÙÙØ´Ø±Ù عÙÙ ÙÙ Ù ÙØ¸Ù.",tokens:"ÙØ¯Ø±Ø© ØÙسبة Ø´ÙØ±ÙØ© Ù Ø´Ù ÙÙØ©. Ø§ÙØ§Ø³ØªÙÙØ§Ù Ø§ÙØ¥Ø¶Ø§ÙÙ Ø¨Ø§ÙØ¯Ùع ØØ³Ø¨ Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù ٠ع ØØ¯ صار٠.",seats:"أعضاء بشرÙÙÙ ÙÙ ÙÙÙÙ Ø§ÙØ¯Ø®ÙÙ ÙØ¥Ø¯Ø§Ø±Ø© ÙÙØ© Ø§ÙØ¹Ù Ù.",api:"ÙÙØ·Ø© ÙÙØ§ÙØ© Ù ØªÙØ§ÙÙØ© ٠ع OpenAI ÙØ§Ø³ØªØ¯Ø¹Ø§Ø¡ Ø§Ø³ØªØ¯ÙØ§Ù Infernex Ù Ù Ø´ÙÙØ±ØªÙ.",analytics:"Ø§ÙØ³Ø§Ø¹Ø§Øª اÙÙ ÙÙÙÙØ±Ø© ÙØªÙÙÙØ© اÙÙ ÙÙ Ø© ÙØ§Ùعائد عÙÙ Ø§ÙØ§Ø³ØªØ«Ù ار Ù٠اÙÙÙØª اÙÙØ¹ÙÙ.",security:"Ø¹Ø²Ù Ø§ÙØ¨ÙØ§ÙØ§Øª ÙØ¶Ùابط اÙÙØµÙÙ ÙØ¶Ù Ø§ÙØ§Øª Ø§ÙØ§Ù تثاÙ.",support:"سرعة ØØµÙÙ٠عÙ٠رد Ø¨Ø´Ø±Ù Ø¹ÙØ¯Ù ا ÙÙÙ Ù Ø§ÙØ£Ù ر."}}}},Fr={en:{cgu:{meta:{title:"Terms of Service â Infernex AI",description:"Terms of Service for the Infernex AI platform: service access, account, acceptable use, billing, liability, termination and applicable law."},badge:"Contractual Document",title:"Terms of Service",subtitle:"These terms govern the access and use of the Infernex AI platform and all its services (AI Workforce, inference API, GEO, IAO Chat).",lastUpdated:"Last updated:",tocTitle:"Table of Contents",contactCta:"Questions about these terms? Write to us at",sections:{definitions:{title:"1. Definitions",content:`âInfernexâ, âweâ: Infernex Corp., a company established in Geneva (Switzerland), publisher of the platform accessible at infernex.ai. 812âClientâ, âyouâ: any natural or legal person with an account on the platform. 813âServicesâ: all functionalities offered by Infernex, including AI Workforce, OpenAI-compatible inference API, AI agents, connectors, website generation, and associated dashboards. 814âGenerated Contentâ: any text, code, image, website, analysis or decision produced by the Services based on your instructions.`},acceptance:{title:"2. Acceptance of terms",content:`The creation of an account, subscription to a plan or any use of the Services signifies full and complete acceptance of these terms. If you do not accept these terms, you must immediately cease using the Services. 815Infernex may modify these terms at any time. Substantial changes will be notified by email at least 15 days before they come into effect. Continued use of the Services after this date constitutes acceptance.`},access:{title:"3. Service access",content:`The Services are provided in SaaS mode, accessible 24/7, subject to maintenance operations and cases of force majeure. 816Infernex implements reasonable means to ensure high availability, without guaranteeing uninterrupted or error-free operation. Contractual service levels (SLA) only apply to plans that expressly mention them.`},account:{title:"4. User account and security",content:`You agree to provide accurate information upon registration and to keep it updated. 817You are solely responsible for the confidentiality of your identifiers, API keys, and access tokens. Any action performed from your account is deemed to have been performed by you. 818You must notify Infernex without delay of any unauthorized access. Sharing an account between multiple distinct entities is prohibited.`},acceptableUse:{title:"5. Acceptable use",content:`It is strictly forbidden to use the Services to: generate or disseminate illegal, hateful, defamatory, deceptive, or child pornography content; violate intellectual property rights; circumvent security measures; perform massive scraping, spam, or identity theft; train a competing model using the outputs of our models; or resell access to the Services without prior written agreement. 819Infernex may immediately suspend any account in violation of this clause without notice.`},plansBilling:{title:"6. Plans, billing and quotas",content:`Prices displayed are exclusive of applicable taxes. Subscriptions are billed
819in advance, monthly or annually depending on the chosen plan, via our payment provider. 820Each plan includes a quota of tokens and features specified on the pricing page. Any consumption beyond the quota is billed at the current overage rate, up to the monthly limit you set. 821âLifetimeâ offers remain valid as long as the subscription is active and paid without interruption. Unless otherwise provided by law, amounts paid are not refundable on a pro rata basis.`},externalCosts:{title:"7. External costs not included",content:`Not included in the plans are: domain names, third-party hosting or messaging, licenses and subscriptions for third-party connectors, your own API keys (BYOK), token overages, taxes and bank fees, and optional services. 822Also not included are the contracts, licenses, and mandatory legal obligations for your company to be registered, compliant, and operational in your country (commercial register, insurance, local taxation, GDPR, professional licenses, etc.): these remain entirely your responsibility. 823The final budget of your company depends on your iterations, the number of active AI employees, and your usage volume.`},ip:{title:"8. Intellectual property",content:`Infernex retains all rights to the platform, its software, models, interfaces, brands, and documentation. 824You retain ownership of the data you submit. To the extent permitted by law, you hold the exploitation rights to the Generated Content based on your instructions, subject to full payment of all sums due. 825You are responsible for verifying that the Generated Content does not infringe any third-party rights before any commercial exploitation.`},data:{title:"9. Personal data and confidentiality",content:`The processing of personal data is described in our Privacy Policy, which forms an integral part of these terms. 826Infernex acts as a data processor for the data you process via the Services and undertakes to comply with the GDPR as well as the announced security standards (encryption in transit, Row-Level Security isolation, sovereign hosting).`},liability:{title:"10. Liability",content:`Any action undertaken on Infernex AI is the sole responsibility of the Client. You are solely responsible for decisions, content, communications, and automations produced or triggered via the platform, as well as their compliance with applicable laws in your country. 827Technical support is available 24/7 by email. Infernex disclaims all liability in case of inappropriate, unlawful, or non-compliant use of the Services by the Client, as well as for indirect damages (loss of turnover, data, opportunity, or image). 828In all cases, Infernex's total liability is limited to the amount actually paid by the Client for the twelve (12) months preceding the generating event.`},ai:{title:"11. Nature of AI-generated results",content:`The Services rely on probabilistic artificial intelligence models. Results may contain inaccuracies, omissions, or biases. 829Generated Content does not constitute legal, tax, medical, financial, or regulatory advice in any way. Human validation is required before any critical use or publication.`},termination:{title:"12. Suspension and termination",content:`You can terminate your subscription at any time from your dashboard; termination takes effect at the end of the current billing period. 830Infernex may suspend or terminate access in case of non-payment, violation of these terms, or use that jeopardizes the security of the platform. 831After termination, your data is kept for 30 days then permanently deleted, unless there is a legal obligation to retain it.`},law:{title:"13. Applicable law and jurisdiction",content:`These terms are governed by Swiss law. In case of dispute, the parties will endeavor to find an amicable solution. 832Failing an agreement, exclusive jurisdiction is attributed to the ordinary courts of the canton of Geneva (Switzerland), subject to mandatory rules protecting consumers.`},contact:{title:"14. Contact",content:`Infernex Corp. â Geneva, Switzerland 833Support and contractual questions: [email protected] 834We respond 24/7 by email.`}}}},fr:{cgu:{meta:{title:"Conditions Générales d'Utilisation â Infernex AI",description:"Conditions Générales d'Utilisation de la plateforme Infernex AI : accès au service, compte, usage acceptable, facturation, responsabilité, résiliation et droit applicable."},badge:"Document contractuel",title:"Conditions Générales d'Utilisation",subtitle:"Les présentes conditions encadrent l'accès et l'utilisation de la plateforme Infernex AI et de l'ensemble de ses services (AI Workforce, API d'inférence, GEO, IAO Chat).",lastUpdated:"Dernière mise à jour :",tocTitle:"Sommaire",contactCta:"Une question sur ces conditions ? Ãcrivez-nous à ",sections:{definitions:{title:"1. Définitions",content:`« Infernex », « nous » : Infernex Corp., société établie à Genève (Suisse), éditrice de la plateforme accessible à l'adresse infernex.ai. 835« Client », « vous » : toute personne physique ou morale disposant d'un compte sur la plateforme. 836« Services » : l'ensemble des fonctionnalités proposées par Infernex, notamment l'AI Workforce, l'API d'inférence compatible OpenAI, les agents IA, les connecteurs, la génération de sites web et les tableaux de bord associés. 837« Contenu généré » : tout texte, code, image, site web, analyse ou décision produit par les Services à partir de vos instructions.`},acceptance:{title:"2. Acceptation des conditions",content:`La création d'un compte, la souscription à un plan ou toute utilisation des Services vaut acceptation pleine et entière des présentes conditions. Si vous n'acceptez pas ces conditions, vous devez cesser immédiatement d'utiliser les Services. 838Infernex peut modifier ces conditions à tout moment. Les modifications substantielles sont notifiées par email au moins 15 jours avant leur entrée en vigueur. La poursuite de l'utilisation des Services après cette date vaut acceptation.`},access:{title:"3. Accès au service",content:`Les Services sont fournis en mode SaaS, accessibles 24h/24 et 7j/7, sous réserve des opérations de maintenance et des cas de force majeure. 839Infernex met en Åuvre les moyens raisonnables pour assurer une haute disponibilité, sans garantie de fonctionnement ininterrompu ou exempt d'erreurs. Les niveaux de service contractuels (SLA)
839ne s'appliquent qu'aux plans qui les mentionnent expressément.`},account:{title:"4. Compte utilisateur et sécurité",content:`Vous vous engagez à fournir des informations exactes lors de l'inscription et à les maintenir à jour. 840Vous êtes seul responsable de la confidentialité de vos identifiants, clés API et jetons d'accès. Toute action réalisée depuis votre compte est réputée effectuée par vous. 841Vous devez notifier Infernex sans délai de tout accès non autorisé. Le partage d'un compte entre plusieurs entités distinctes est interdit.`},acceptableUse:{title:"5. Usage acceptable",content:`Il est strictement interdit d'utiliser les Services pour : générer ou diffuser des contenus illégaux, haineux, diffamatoires, trompeurs ou pédopornographiques ; violer des droits de propriété intellectuelle ; contourner des mesures de sécurité ; procéder à du scraping massif, du spam ou de l'usurpation d'identité ; entraîner un modèle concurrent à partir des sorties de nos modèles ; ou revendre l'accès aux Services sans accord écrit préalable. 842Infernex peut suspendre immédiatement et sans préavis tout compte en violation de cette clause.`},plansBilling:{title:"6. Plans, facturation et quotas",content:`Les prix affichés sont hors taxes applicables. Les abonnements sont facturés d'avance, mensuellement ou annuellement selon le plan choisi, via notre prestataire de paiement. 843Chaque plan inclut un quota de jetons et de fonctionnalités précisé sur la page tarifaire. Toute consommation au-delà du quota est facturée au tarif de dépassement en vigueur, dans la limite du plafond mensuel que vous définissez. 844Les offres « à vie » restent valables tant que l'abonnement est actif et payé sans interruption. Sauf disposition légale contraire, les sommes versées ne sont pas remboursables au prorata.`},externalCosts:{title:"7. Frais externes non inclus",content:`Ne sont pas inclus dans les plans : les noms de domaine, l'hébergement ou la messagerie tiers, les licences et abonnements de connecteurs tiers, vos propres clés API (BYOK), les dépassements de jetons, les taxes et frais bancaires, ainsi que les prestations optionnelles. 845Ne sont pas non plus inclus les contrats, licences et obligations légales obligatoires pour que votre entreprise soit immatriculée, conforme et opérationnelle dans votre pays (registre du commerce, assurances, fiscalité locale, RGPD, licences professionnelles, etc.) : ils restent entièrement à votre charge. 846Le budget final de votre entreprise dépend de vos itérations, du nombre d'employés IA actifs et de votre volume d'usage.`},ip:{title:"8. Propriété intellectuelle",content:`Infernex conserve l'intégralité des droits sur la plateforme, ses logiciels, modèles, interfaces, marques et documentations. 847Vous conservez la propriété des données que vous soumettez. Dans les limites permises par la loi, vous détenez les droits d'exploitation du Contenu généré à partir de vos instructions, sous réserve du paiement intégral des sommes dues. 848Vous êtes responsable de vérifier que le Contenu généré n'enfreint aucun droit de tiers avant toute exploitation commerciale.`},data:{title:"9. Données personnelles et confidentialité",content:`Le traitement des données personnelles est décrit dans notre Politique de Confidentialité, qui fait partie intégrante des présentes conditions. 849Infernex agit en qualité de sous-traitant pour les données que vous traitez via les Services et s'engage à respecter le RGPD ainsi que les standards de sécurité annoncés (chiffrement en transit, isolation par Row-Level Security, hébergement souverain).`},liability:{title:"10. Responsabilité",content:`Toute action entreprise sur Infernex AI relève de la seule responsabilité du Client. Vous êtes seul responsable des décisions, contenus, communications et automatisations produits ou déclenchés via la plateforme, ainsi que de leur conformité aux lois applicables dans votre pays. 850Un support technique est disponible 24h/24 par email. Infernex décline toute responsabilité en cas d'usage inapproprié, illicite ou non conforme des Services par le Client, ainsi que pour les dommages indirects (perte de chiffre d'affaires, de données, d'opportunité ou d'image). 851Dans tous les cas, la responsabilité globale d'Infernex est limitée au montant effectivement payé par le Client au titre des douze (12) derniers mois précédant le fait générateur.`},ai:{title:"11. Nature des résultats générés par l'IA",content:`Les Services reposent sur des modèles d'intelligence artificielle probabilistes. Les résultats peuvent contenir des inexactitudes, des omissions ou des biais. 852Le Contenu généré ne constitue en aucun cas un conseil juridique, fiscal, médical, financier ou réglementaire. Une validation humaine est requise avant toute utilisation critique ou publication.`},termination:{title:"12. Suspension et résiliation",content:`Vous pouvez résilier votre abonnement à tout moment depuis votre tableau de bord ; la résiliation prend effet à la fin de la période de facturation en cours.
853Infernex peut suspendre ou résilier l'accès en cas de non-paiement, de violation des présentes conditions ou d'usage mettant en péril la sécurité de la plateforme. 854Après résiliation, vos données sont conservées 30 jours puis supprimées définitivement, sauf obligation légale de conservation.`},law:{title:"13. Droit applicable et juridiction",content:`Les présentes conditions sont régies par le droit suisse. En cas de litige, les parties s'efforceront de trouver une solution amiable. 855à défaut d'accord, compétence exclusive est attribuée aux tribunaux ordinaires du canton de Genève (Suisse), sous réserve des règles impératives protégeant les consommateurs.`},contact:{title:"14. Contact",content:`Infernex Corp. â Genève, Suisse 856Support et questions contractuelles : [email protected] 857Nous répondons 24h/24 par email.`}}}},es:{cgu:{meta:{title:"Condiciones Generales de Uso â Infernex AI",description:"Condiciones Generales de Uso de la plataforma Infernex AI: acceso al servicio, cuenta, uso aceptable, facturación, responsabilidad, terminación y ley aplicable."},badge:"Documento contractual",title:"Condiciones Generales de Uso",subtitle:"Las presentes condiciones rigen el acceso y uso de la plataforma Infernex AI y de todos sus servicios (AI Workforce, API de inferencia, GEO, IAO Chat).",lastUpdated:"Ãltima actualización:",tocTitle:"Ãndice",contactCta:"¿Preguntas sobre estas condiciones? EscrÃbanos a",sections:{definitions:{title:"1. Definiciones",content:`« Infernex », « nosotros »: Infernex Corp., empresa establecida en Ginebra (Suiza), editora de la plataforma accesible en infernex.ai. 858« Cliente », « usted »: cualquier persona fÃsica o jurÃdica que tenga una cuenta en la plataforma. 859« Servicios »: todas las funcionalidades ofrecidas por Infernex, incluyendo AI Workforce, la API de inferencia compatible con OpenAI, los agentes de IA, los conectores, la generación de sitios web y los paneles de control asociados. 860« Contenido generado »: cualquier texto, código, imagen, sitio web, análisis o decisión producido por los Servicios a partir de sus instrucciones.`},acceptance:{title:"2. Aceptación de las condiciones",content:`La creación de una cuenta, la suscripción a un plan o cualquier uso de los Servicios implica la aceptación total de las presentes condiciones. Si no acepta estas condiciones, debe dejar de usar los Servicios inmediatamente. 861Infernex puede modificar estas condiciones en cualquier momento. Las modificaciones sustanciales se notificarán por correo electrónico al menos 15 dÃas antes de su entrada en vigor. El uso continuado de los Servicios después de esta fecha constituye la aceptación de las mismas.`},access:{title:"3. Acceso al servicio",content:`Los Servicios se proporcionan en modo SaaS, accesibles 24 horas al dÃa, 7 dÃas a la semana, sujetos a operaciones de mantenimiento y casos de fuerza mayor. 862Infernex implementa medios razonables para asegurar una alta disponibilidad, sin garantÃa de funcionamiento ininterrumpido o libre de errores. Los niveles de servicio contractuales (SLA) solo se aplican a los planes que los mencionan expresamente.`},account:{title:"4. Cuenta de usuario y seguridad",content:`Usted se compromete a proporcionar información precisa al registrarse y a mantenerla actualizada. 863Usted es el único responsable de la confidencialidad de sus credenciales, claves API y tokens de acceso. Cualquier acción realizada desde su cuenta se considera realizada por usted. 864Debe notificar a Infernex sin demora cualquier acceso no autorizado. Está prohibido compartir una cuenta entre varias entidades distintas.`},acceptableUse:{title:"5. Uso aceptable",content:`Está estrictamente prohibido utilizar los Servicios para: generar o difundir contenido ilegal, odioso, difamatorio, engañoso o pedopornográfico; violar derechos de propiedad intelectual; eludir medidas de seguridad; realizar scraping masivo, spam o suplantación de identidad; entrenar un modelo que compita a partir de los resultados de nuestros modelos; o revender el acceso a los Servicios sin acuerdo previo por escrito. 865Infernex puede suspender de inmediato y sin previo aviso cualquier cuenta que infrinja esta cláusula.`},plansBilling:{title:"6. Planes, facturación y cuotas",content:`Los precios mostrados no incluyen los impuestos aplicables. Las suscripciones se facturan por adelantado, mensual o anualmente según el plan elegido, a través de nuestro proveedor de pagos. 866Cada plan incluye una cuota de tokens y funcionalidades especificadas en la página de precios. Cualquier consumo más allá de la cuota se facturará a la tarifa de exceso vigente, dentro del lÃmite mensual que usted defina. 867Las ofertas «de por vida» seguirán siendo válidas mientras la suscripción esté activa y pagada sin interrupción. Salvo disposición legal en contrario, las sumas pagadas no son reembolsables a prorrata.`},externalCosts:{title:"7. Gastos externos no incluidos",content:`No se incluyen en los planes: nombres de dominio, alojamiento o correo electrónico de terceros, licencias y suscripciones de conectores de terceros, sus propias claves API (BYOK), excedentes de tokens, impuestos y tarifas bancarias, asà como los servicios opcionales. 868Tampoco se incluyen los contratos, licencias y obligaciones legales obligatorios para que su empresa esté registrada, conforme y operativa en su paÃs (registro mercantil, seguros, impuestos locales, RGPD, licencias profesionales, etc.): estos correrán Ãntegramente por su cuenta.
869El presupuesto final de su empresa depende de sus interacciones, el número de empleados de IA activos y su volumen de uso.`},ip:{title:"8. Propiedad intelectual",content:`Infernex conserva todos los derechos sobre la plataforma, su software, modelos, interfaces, marcas y documentación. 870Usted mantiene la propiedad de los datos que envÃa. En la medida permitida por la ley, usted posee los derechos de explotación del Contenido generado a partir de sus instrucciones, sujeto al pago Ãntegro de las sumas adeudadas. 871Usted es responsable de verificar que el Contenido generado no infringe ningún derecho de terceros antes de cualquier explotación comercial.`},data:{title:"9. Datos personales y confidencialidad",content:`El tratamiento de datos personales se describe en nuestra PolÃtica de Privacidad, que forma parte integrante de las presentes condiciones. 872Infernex actúa como encargado del tratamiento para los datos que usted procesa a través de los Servicios y se compromete a cumplir con el RGPD, asà como con los estándares de seguridad anunciados (cifrado en tránsito, aislamiento por seguridad a nivel de fila, alojamiento soberano).`},liability:{title:"10. Responsabilidad",content:`Toda acción realizada en Infernex AI es responsabilidad exclusiva del Cliente. Usted es el único responsable de las decisiones, contenidos, comunicaciones y automatizaciones producidas o activadas a través de la plataforma, asà como de su conformidad con las leyes aplicables en su paÃs. 873Hay soporte técnico disponible las 24 horas del dÃa por correo electrónico. Infernex declina toda responsabilidad en caso de uso inapropiado, ilÃcito o no conforme de los Servicios por parte del Cliente, asà como por los daños indirectos (pérdida de facturación, de datos, de oportunidad o de imagen). 874En todos los casos, la responsabilidad global de Infernex se limita al importe efectivamente pagado por el Cliente durante los doce (12) meses anteriores al hecho causante.`},ai:{title:"11. Naturaleza de los resultados generados por la IA",content:`Los Servicios se basan en modelos de inteligencia artificial probabilÃsticos. Los resultados pueden contener inexactitudes, omisiones o sesgos. 875El Contenido generado no constituye en ningún caso un consejo legal, fiscal, médico, financiero o reglamentario. Se requiere una validación humana antes de cualquier uso crÃtico o publicación.`},termination:{title:"12. Suspensión y terminación",content:`Puede cancelar su suscripción en cualquier momento desde su panel de control; la cancelación entrará en vigor al final del perÃodo de facturación actual. 876Infernex puede suspender o cancelar el acceso en caso de impago, incumplimiento de estas condiciones o uso que ponga en peligro la seguridad de la plataforma. 877Después de la terminación, sus datos se conservarán durante 30 dÃas y luego se eliminarán permanentemente, salvo obligación legal de conservación.`},law:{title:"13. Ley aplicable y jurisdicción",content:`Estas condiciones se rigen por la ley suiza. En caso de disputa, las partes se esforzarán por encontrar una solución amistosa. 878A falta de acuerdo, la competencia exclusiva se atribuye a los tribunales ordinarios del cantón de Ginebra (Suiza), sin perjuicio de las normas imperativas que protegen a los consumidores.`},contact:{title:"14. Contacto",content:`Infernex Corp. â Ginebra, Suiza 879Soporte y preguntas contractuales: [email protected] 880Respondemos 24 horas al dÃa por correo electrónico.`}}}},de:{cgu:{meta:{title:"Allgemeine Geschäftsbedingungen â Infernex AI",description:"Allgemeine Geschäftsbedingungen der Infernex AI-Plattform: Servicezugang, Konto, akzeptable Nutzung, Abrechnung, Haftung, Kündigung und anwendbares Recht."},badge:"Vertragsdokument",title:"Allgemeine Geschäftsbedingungen",subtitle:"Diese Bedingungen regeln den Zugang und die Nutzung der Infernex AI-Plattform und all ihrer Dienste (AI Workforce, Inferenz-API, GEO, IAO Chat).",lastUpdated:"Letzte Aktualisierung:",tocTitle:"Inhaltsverzeichnis",contactCta:"Fragen zu diesen Bedingungen? Schreiben Sie uns an",sections:{definitions:{title:"1. Definitionen",content:`âInfernexâ, âwirâ: Infernex Corp., ein in Genf (Schweiz) ansässiges Unternehmen, Betreiber der Plattform, die unter infernex.ai zugänglich ist. 881âKundeâ, âSieâ: jede natürliche oder juristische Person mit einem Konto auf der Plattform. 882âDiensteâ: alle von Infernex angebotenen Funktionen, insbesondere AI Workforce, OpenAI-kompatible Inferenz-API, KI-Agenten, Konnektoren, Website-Generierung und zugehörige Dashboards. 883âGenerierter Inhaltâ: jeder Text, Code, Bild, jede Website, Analyse oder Entscheidung, die von den Diensten auf der Grundlage Ihrer Anweisungen erstellt wurde.`},acceptance:{title:"2. Akzeptanz der Bedingungen",content:`Durch die Erstellung eines Kontos, den Abschluss eines Plans oder die Nutzung der Dienste stimmen Sie diesen Bedingungen vollumfänglich zu. Wenn Sie diesen Bedingungen nicht zustimmen, müssen Sie die Nutzung der Dienste sofort einstellen. 884Infernex kann diese Bedingungen jederzeit ändern. Wesentliche Ãnderungen werden Ihnen mindestens 15 Tage vor Inkrafttreten per E-Mail mitgeteilt. Die fortgesetzte Nutzung der Dienste nach diesem Datum gilt als Akzeptanz.`},access:{title:"3. Zugang zum Dienst",content:`Die Dienste werden im SaaS-Modus bereitgestellt, rund um die Uhr zugänglich, vorbehaltlich Wartungsarbeiten und Fällen höherer Gewalt. 885Infernex unternimmt angemessene Anstrengungen, um eine hohe Verfügbarkeit zu gewährleisten, ohne jedoch einen unterbrechungsfreien oder fehlerfreien Betrieb zu garantieren. Vertragliche Service Level Agreements (SLAs) gelten nur für Pläne, die diese ausdrücklich erwähnen.`},account:{title:"4. Benutzerkonto und Sicherheit",content:`Sie verpflichten sich, bei der Registrierung genaue Informationen anzugeben und diese aktuell zu halten. 886Sie sind allein verantwortlich für die Vertraulichkeit Ihrer Anmeldeinformationen, API-Schlüssel und Zugriffstoken. Jede von Ihrem Konto aus durchgeführte Aktion gilt als von Ihnen ausgeführt. 887Sie müssen Infernex unverzüglich über jeden unbefugten Zugriff informieren. Die gemeinsame Nutzung eines Kontos durch mehrere separate Einheiten ist untersagt.`},acceptableUse:{title:"5. Akzeptable Nutzung",content:`Es ist strengstens untersagt, die Dienste zu verwenden, um: illegale, hasserfüllte, diffamierende, irreführende oder kinderpornografische Inhalte zu generieren oder zu verbreiten; geistige Eigentumsrechte zu verletzen; SicherheitsmaÃnahmen zu umgehen; massives Scraping, Spam oder Identitätsdiebstahl durchzuführen; ein konkurrierendes Modell aus den Ausgaben unserer Modelle zu trainieren; oder den Zugang zu den Diensten ohne vorherige schriftliche Zustimmung weiterzuverkaufen. 888Infernex kann Konten, die gegen diese Klausel verstoÃen, sofort und ohne Vorankündigung sperren.`},plansBilling:{title:"6. Pläne, Abrechnung und Quoten",content:`Die angezeigten Preise verstehe
888n sich zuzüglich anfallender Steuern. Abonnements werden im Voraus, monatlich oder jährlich je nach gewähltem Plan, über unseren Zahlungsdienstleister abgerechnet. 889Jeder Plan beinhaltet eine Quote für Token und Funktionen, die auf der Preisseite aufgeführt ist. Jeder Verbrauch über die Quote hinaus wird zum geltenden Ãberziehungstarif abgerechnet, innerhalb der von Ihnen festgelegten monatlichen Obergrenze. 890âLebenslangeâ Angebote bleiben gültig, solange das Abonnement aktiv ist und ununterbrochen bezahlt wird. Sofern gesetzlich nichts anderes vorgeschrieben ist, sind die gezahlten Beträge nicht anteilig erstattungsfähig.`},externalCosts:{title:"7. Nicht enthaltene externe Kosten",content:`Nicht in den Plänen enthalten sind: Domainnamen, Hosting oder Drittanbieter-E-Mail, Lizenzen und Abonnements von Drittanbieter-Konnektoren, Ihre eigenen API-Schlüssel (BYOK), Token-Ãberschreitungen, Steuern und Bankgebühren sowie optionale Dienstleistungen. 891Ebenfalls nicht enthalten sind Verträge, Lizenzen und gesetzliche Verpflichtungen, die für die Registrierung, Konformität und den Betrieb Ihres Unternehmens in Ihrem Land erforderlich sind (Handelsregister, Versicherungen, lokale Steuern, DSGVO, BerufsLizenzen usw.): diese bleiben vollständig in Ihrer Verantwortung. 892Das Endbudget Ihres Unternehmens hängt von Ihren Iterationen, der Anzahl der aktiven KI-Mitarbeiter und Ihrem Nutzungsvolumen ab.`},ip:{title:"8. Geistiges Eigentum",content:`Infernex behält alle Rechte an der Plattform, ihrer Software, Modellen, Schnittstellen, Marken und Dokumentationen. 893Sie behalten das Eigentum an den von Ihnen übermittelten Daten. Im gesetzlich zulässigen Rahmen besitzen Sie die Nutzungsrechte an den generierten Inhalten auf der Grundlage Ihrer Anweisungen, vorbehaltlich der vollständigen Zahlung der fälligen Beträge. 894Sie sind verantwortlich dafür, vor jeder kommerziellen Nutzung zu überprüfen, ob der generierte Inhalt keine Rechte Dritter verletzt.`},data:{title:"9. Persönliche Daten und Vertraulichkeit",content:`Die Verarbeitung personenbezogener Daten wird in unserer Datenschutzerklärung beschrieben, die integraler Bestandteil dieser Bedingungen ist. 895Infernex fungiert als Auftragsverarbeiter für die Daten, die Sie über die Dienste verarbeiten, und verpflichtet sich zur Einhaltung der DSGVO sowie der angekündigten Sicherheitsstandards (Verschlüsselung im Transit, Isolation durch Row-Level Security, souveränes Hosting).`},liability:{title:"10. Haftung",content:`Jede Handlung, die auf Infernex AI durchgeführt wird, liegt in der alleinigen Verantwortung des Kunden. Sie sind allein verantwortlich für die Entscheidungen, Inhalte, Kommunikationen und Automatisierungen, die über die Plattform erstellt oder ausgelöst werden, sowie für deren Einhaltung der in Ihrem Land geltenden Gesetze. 896Technischer Support ist 24 Stunden am Tag per E-Mail verfügbar. Infernex lehnt jede Verantwortung für eine unangemessene, unerlaubte oder nicht konforme Nutzung der Dienste durch den Kunden sowie für indirekte Schäden (Verlust von Umsatz, Daten, Gelegenheiten oder Image) ab. 897In jedem Fall ist die Gesamthaftung von Infernex auf den Betrag begrenzt, der vom Kunden in den letzten zwölf (12) Monaten vor dem auslösenden Ereignis tatsächlich bezahlt wurde.`},ai:{title:"11. Art der von der KI generierten Ergebnisse",content:`Die Dienste basieren auf probabilistischen künstlichen Intelligenzmodellen. Die Ergebnisse können Ungenauigkeiten, Auslassungen oder Verzerrungen enthalten. 898Der generierte Inhalt stellt in keinem Fall eine rechtliche, steuerliche, medizinische, finanzielle oder regulatorische Beratung dar. Eine menschliche Validierung ist vor jeder kritischen Nutzung oder Veröffentlichung erforderlich.`},termination:{title:"12. Sperrung und Kündigung",content:`Sie können Ihr Abonnement jederzeit über Ihr Dashboard kündigen; die Kündigung wird zum Ende des laufenden Abrechnungszeitraums wirksam. 899Infernex kann den Zugang im Falle von Nichtzahlung, Verletzung dieser Bedingungen oder einer Nutzung, die die Sicherheit der Plattform gefährdet, sperren oder kündigen. 900Nach der Kündigung werden Ihre Daten 30 Tage lang aufbewahrt und dann endgültig gelöscht, es sei denn, es besteht eine gesetzliche Aufbewahrungspflicht.`},law:{title:"13. Anwendbares Recht und Gerichtsstand",content:`Diese Bedingungen unterliegen schweizerischem Recht. Im Falle eines Rechtsstreits werden die Parteien versuchen, eine gütliche Lösung zu finden. 901Bei fehlender Einigung wird die ausschlieÃliche Zuständigkeit den ordentlichen Gerichten des Kantons Genf (Schweiz) zugeschrieben, vorbehaltlich zwingender Bestimmungen zum Schutz der Verbraucher.`},contact:{title:"14. Kontakt",content:`Infernex Corp. â Genf, Schweiz 902Support und vertragliche Fragen: [email protected] 903Wir antworten 24 Stunden am Tag per E-Mail.`}}}},pt:{cgu:{meta:{title:"Termos de Serviço â Infernex AI",description:"Termos de Serviço da plataforma Infernex AI: acesso ao serviço, conta, uso ace
903itável, faturamento, responsabilidade, rescisão e lei aplicável."},badge:"Documento contratual",title:"Termos de Serviço",subtitle:"Estes termos regem o acesso e uso da plataforma Infernex AI e todos os seus serviços (AI Workforce, API de inferência, GEO, IAO Chat).",lastUpdated:"Ãltima atualização:",tocTitle:"Sumário",contactCta:"Dúvidas sobre estes termos? Escreva para",sections:{definitions:{title:"1. Definições",content:`« Infernex », « nós »: Infernex Corp., empresa estabelecida em Genebra (SuÃça), editora da plataforma acessÃvel em infernex.ai. 904« Cliente », « você »: qualquer pessoa fÃsica ou jurÃdica com uma conta na plataforma. 905« Serviços »: todas as funcionalidades oferecidas pela Infernex, incluindo AI Workforce, API de inferência compatÃvel com OpenAI, agentes de IA, conectores, geração de sites e painéis associados. 906« Conteúdo Gerado »: qualquer texto, código, imagem, site, análise ou decisão produzido pelos Serviços com base em suas instruções.`},acceptance:{title:"2. Aceitação dos termos",content:`A criação de uma conta, a assinatura de um plano ou qualquer uso dos Serviços constitui aceitação plena e total destes termos. Se você não concorda com estes termos, deve parar imediatamente de usar os Serviços. 907A Infernex pode modificar estes termos a qualquer momento. Modificações substanciais serão notificadas por e-mail com pelo menos 15 dias de antecedência antes de sua entrada em vigor. A continuação do uso dos Serviços após esta data constitui aceitação.`},access:{title:"3. Acesso ao serviço",content:`Os Serviços são fornecidos em modo SaaS, acessÃveis 24 horas por dia, 7 dias por semana, sujeito a operações de manutenção e casos de força maior. 908A Infernex emprega meios razoáveis para garantir alta disponibilidade, sem garantia de operação ininterrupta ou livre de erros. Os nÃveis de serviço contratuais (SLA) aplicam-se apenas aos planos que os mencionam expressamente.`},account:{title:"4. Conta de usuário e segurança",content:`Você se compromete a fornecer informações exatas ao se registrar e a mantê-las atualizadas. 909Você é o único responsável pela confidencialidade de suas credenciais, chaves de API e tokens de acesso. Qualquer ação realizada a partir de sua conta é considerada feita por você. 910Você deve notificar a Infernex sem demora sobre qualquer acesso não autorizado. O compartilhamento de uma conta entre várias entidades distintas é proibido.`},acceptableUse:{title:"5. Uso aceitável",content:`à estritamente proibido usar os Serviços para: gerar ou disseminar conteúdo ilegal, odioso, difamatório, enganoso ou pedopornográfico; violar direitos de propriedade intelectual; contornar medidas de segurança; realizar scraping massivo, spam ou usurpação de identidade; treinar um modelo concorrente a partir das saÃdas de nossos modelos; ou revender o acesso aos Serviços sem acordo escrito prévio. 911A Infernex pode suspender imediatamente e sem aviso prévio qualquer conta em violação desta cláusula.`},plansBilling:{title:"6. Planos, faturamento e cotas",content:`Os preços exibidos são exclusivos de impostos aplicáveis. As assinaturas são cobradas antecipadamente, mensalmente ou anualmente, dependendo do plano escolhido, através do nosso provedor de pagamento. 912Cada plano inclui uma cota de tokens e funcionalidades especificadas na página de preços. Qualquer consumo além da cota é cobrado à tarifa de excedente vigente, dentro do limite mensal que você definir. 913As ofertas âvitalÃciasâ permanecem válidas enquanto a assinatura estiver ativa e paga sem interrupção. Salvo disposição legal em contrário, os valores pagos não são reembolsáveis proporcionalmente.`},externalCosts:{title:"7. Custos externos não incluÃdos",content:`Não estão incluÃdos nos planos: nomes de domÃnio, hospedagem ou e-mail de terceiros, licenças e assinaturas de conectores de terceiros, suas próprias chaves de API (BYOK), excedentes de tokens, impostos e taxas bancárias, bem como serviços opcionais. 914Também não estão incluÃdos os contratos, licenças e obrigações legais obrigatórias para que sua empresa seja registrada, esteja em conformidade e operacional em seu paÃs (registro comercial, seguros, tributação local, GDPR, licenças profissionais, etc.): eles permanecem inteiramente sob sua responsabilidade.
915O orçamento final de sua empresa depende de suas iterações, do número de funcionários de IA ativos e do seu volume de uso.`},ip:{title:"8. Propriedade intelectual",content:`A Infernex retém todos os direitos sobre a plataforma, seus softwares, modelos, interfaces, marcas e documentações. 916Você mantém a propriedade dos dados que envia. Dentro dos limites permitidos por lei, você detém os direitos de uso do Conteúdo Gerado a partir de suas instruções, sujeito ao pagamento integral dos valores devidos. 917Você é responsável por verificar se o Conteúdo Gerado não infringe nenhum direito de terceiros antes de qualquer exploração comercial.`},data:{title:"9. Dados pessoais e confidencialidade",content:`O tratamento de dados pessoais é descrito em nossa PolÃtica de Privacidade, que faz parte integrante destes termos. 918A Infernex atua como processador de dados que você processa por meio dos Serviços e se compromete a cumprir o GDPR, bem como os padrões de segurança anunciados (criptografia em trânsito, isolamento por Row-Level Security, hospedagem soberana).`},liability:{title:"10. Responsabilidade",content:`Qualquer ação empreendida na Infernex AI é de responsabilidade exclusiva do Cliente. Você é o único responsável pelas decisões, conteúdos, comunicações e automações produzidas ou acionadas através da plataforma, bem como pela sua conformidade com as leis aplicáveis em seu paÃs. 919Suporte técnico está disponÃvel 24h/24 por e-mail. A Infernex declina qualquer responsabilidade em caso de uso inadequado, ilÃcito ou não conforme dos Serviços pelo Cliente, bem como por danos indiretos (perda de faturamento, de dados, de oportunidade ou de imagem). 920Em todos os casos, a responsabilidade global da Infernex é limitada ao valor efetivamente pago pelo Cliente nos doze (12) meses anteriores ao evento causador.`},ai:{title:"11. Natureza dos resultados gerados pela IA",content:`Os Serviços são baseados em modelos de inteligência artificial probabilÃsticos. Os resultados podem conter imprecisões, omissões ou vieses. 921O Conteúdo Gerado não constitui em nenhum caso um conselho jurÃdico, fiscal, médico, financeiro ou regulatório. Uma validação humana é necessária antes de qualquer uso crÃtico ou publicação.`},termination:{title:"12. Suspensão e rescisão",content:`Você pode rescindir sua assinatura a qualquer momento a partir do seu painel de controle; a rescisão entra em vigor ao final do perÃodo de faturamento atual. 922A Infernex pode suspender ou rescindir o acesso em caso de não pagamento, violação destes termos ou uso que coloque em risco a segurança da plataforma. 923Após a rescisão, seus dados são retidos por 30 dias e, em seguida, excluÃdos permanentemente, salvo obrigação legal de retenção.`},law:{title:"13. Lei aplicável e jurisdição",content:`Estes termos são regidos pelo direito suÃço. Em caso de litÃgio, as partes se esforçarão para encontrar uma solução amigável. 924Na falta de acordo, a competência exclusiva é atribuÃda aos tribunais ordinários do cantão de Genebra (SuÃça), ressalvando-se as regras imperativas que protegem os consumidores.`},contact:{title:"14. Contato",content:`Infernex Corp. â Genebra, SuÃça 925Suporte e questões contratuais: [email protected] 926Respondemos 24h/24 por e-mail.`}}}},ar:{cgu:{meta:{title:"Ø´Ø±ÙØ· Ø§ÙØ®Ø¯Ù Ø© â Infernex AI",description:"Ø´Ø±ÙØ· Ø§ÙØ®Ø¯Ù Ø© ÙÙ ÙØµØ© Infernex AI: اÙÙØµÙ٠إÙÙ Ø§ÙØ®Ø¯Ù Ø©Ø Ø§ÙØØ³Ø§Ø¨Ø Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù اÙÙ ÙØ¨ÙÙØ اÙÙÙØªØ±Ø©Ø اÙ٠سؤÙÙÙØ©Ø Ø§ÙØ¥ÙÙØ§Ø¡Ø ÙØ§ÙÙØ§ÙÙ٠اÙ٠ع٠Ù٠بÙ."},badge:"ÙØ«ÙÙØ© ØªØ¹Ø§ÙØ¯ÙØ©",title:"Ø´Ø±ÙØ· Ø§ÙØ®Ø¯Ù Ø©",subtitle:"ØªØØ¯Ø¯ ÙØ°Ù Ø§ÙØ´Ø±ÙØ· اÙÙØµÙ٠إÙÙ ÙØ§Ø³ØªØ®Ø¯Ø§Ù Ù ÙØµØ© Infernex AI ÙØ¬Ù ÙØ¹ Ø®Ø¯Ù Ø§ØªÙØ§ (AI Workforce, ÙØ§Ø¬ÙØ© بر٠جة تطبÙÙØ§Øª Ø§ÙØ§Ø³ØªØ¯ÙØ§ÙØ GEO, IAO Chat).",lastUpdated:"آخر ØªØØ¯ÙØ«:",tocTitle:"اÙÙ ØØªÙÙØ§Øª",contactCta:"ÙØ¯Ù٠سؤا٠ØÙÙ ÙØ°Ù Ø§ÙØ´Ø±ÙØ·Ø Ø±Ø§Ø³ÙÙØ§ عÙÙ",sections:{definitions:{title:"1. تعرÙÙØ§Øª",content:`« Infernex Â»Ø Â« ÙØÙ Â»: Infernex Corp.Ø Ø´Ø±ÙØ© تأسست Ù٠جÙÙÙ (سÙÙØ³Ø±Ø§)Ø ÙÙÙ ÙØ§Ø´Ø± اÙÙ ÙØµØ© اÙÙ ØªØ§ØØ© عÙÙ infernex.ai. 927« Ø§ÙØ¹Ù ÙÙ Â»Ø Â« Ø£ÙØªÙ »: أ٠شخص Ø·Ø¨ÙØ¹Ù أ٠اعتبار٠Ù٠تÙÙ ØØ³Ø§Ø¨Ùا عÙ٠اÙÙ ÙØµØ©. 928« Ø§ÙØ®Ø¯Ù ات »: Ø¬Ù ÙØ¹ اÙÙØ¸Ø§Ø¦Ù Ø§ÙØªÙ ØªÙØ¯Ù ÙØ§ InfernexØ Ø¨Ù Ø§ Ù٠ذÙÙ AI WorkforceØ ÙØ§Ø¬ÙØ© بر٠جة تطبÙÙØ§Øª Ø§ÙØ§Ø³ØªØ¯Ùا٠اÙÙ ØªÙØ§ÙÙØ© ٠ع OpenAIØ ÙÙÙØ§Ø¡ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹ÙØ اÙÙ ÙØµÙØ§ØªØ Ø¥ÙØ´Ø§Ø¡ Ù ÙØ§Ùع اÙÙÙØ¨Ø ÙÙÙØØ§Øª Ø§ÙØªØÙ٠اÙ٠رتبطة. 929« اÙÙ ØØªÙ٠اÙÙ ÙÙØ´Ø£ »: Ø£Ù ÙØµØ Ø±Ù Ø²Ø ØµÙØ±Ø©Ø Ù ÙÙØ¹ ÙÙØ¨Ø تØÙÙÙØ Ø£Ù ÙØ±Ø§Ø± ØªÙØªØ¬Ù Ø§ÙØ®Ø¯Ù ات Ø¨ÙØ§Ø¡Ù عÙ٠تعÙÙ٠اتÙÙ .`},acceptance:{title:"2. ÙØ¨ÙÙ Ø§ÙØ´Ø±ÙØ·",content:`Ø¥Ù Ø¥ÙØ´Ø§Ø¡ ØØ³Ø§Ø¨Ø Ø£Ù Ø§ÙØ§Ø´ØªØ±Ø§Ù ÙÙ Ø®Ø·Ø©Ø Ø£Ù Ø£Ù Ø§Ø³ØªØ®Ø¯Ø§Ù ÙÙØ®Ø¯Ù ات ÙØ¹Ù٠اÙÙØ¨Ù٠اÙÙØ§Ù Ù ÙÙØ°Ù Ø§ÙØ´Ø±ÙØ·. إذا ÙÙ ØªÙØ§Ù٠عÙÙ ÙØ°Ù Ø§ÙØ´Ø±ÙØ·Ø ÙØ¬Ø¨ عÙÙÙ Ø§ÙØªÙÙÙ ÙÙØ±Ùا Ø¹Ù Ø§Ø³ØªØ®Ø¯Ø§Ù Ø§ÙØ®Ø¯Ù ات. 930ÙÙ ÙÙ ÙÙ Infernex تعدÙÙ ÙØ°Ù Ø§ÙØ´Ø±ÙØ· Ù٠أ٠ÙÙØª. ÙØªÙ إخطار Ø§ÙØªØ¹Ø¯ÙÙØ§Øª Ø§ÙØ¬ÙÙØ±ÙØ© عبر Ø§ÙØ¨Ø±Ùد Ø§ÙØ¥ÙÙØªØ±ÙÙÙ ÙØ¨Ù 15 ÙÙÙ ÙØ§ عÙÙ Ø§ÙØ£Ù٠٠٠دخÙÙÙØ§ ØÙز Ø§ÙØªÙÙÙØ°. است٠رار Ø§Ø³ØªØ®Ø¯Ø§Ù Ø§ÙØ®Ø¯Ù ات بعد ÙØ°Ø§ Ø§ÙØªØ§Ø±ÙØ® ÙØ¹ØªØ¨Ø± ÙØ¨ÙÙÙØ§.`},access:{title:"3. اÙÙØµÙ٠إÙÙ Ø§ÙØ®Ø¯Ù Ø©",content:`ÙØªÙ تÙÙÙØ± Ø§ÙØ®Ø¯Ù ات ÙØ¨Ø±Ùا٠ج ÙØ®Ø¯Ù Ø© (SaaS)Ø ÙÙÙ Ù٠اÙÙØµÙ٠إÙÙÙØ§ 24 ساعة Ù٠اÙÙÙÙ Ø 7 Ø£ÙØ§Ù ÙÙ Ø§ÙØ£Ø³Ø¨ÙØ¹Ø Ø¨Ø§Ø³ØªØ«ÙØ§Ø¡ ع٠ÙÙØ§Øª Ø§ÙØµÙØ§ÙØ© ÙØØ§ÙØ§Øª اÙÙÙØ© اÙÙØ§Ùرة. 931تبذ٠Infernex جÙÙØ¯Ùا ٠عÙÙÙØ© ÙØ¶Ù ا٠تÙÙØ± عاÙÙØ دÙÙ Ø¶Ù Ø§Ù Ø§ÙØªØ´ØºÙ٠اÙ٠ست٠ر Ø£Ù Ø§ÙØ®Ø§ÙÙ Ù Ù Ø§ÙØ£Ø®Ø·Ø§Ø¡. ÙØ§ ØªÙØ·Ø¨Ù ٠ستÙÙØ§Øª Ø§ÙØ®Ø¯Ù Ø© Ø§ÙØªØ¹Ø§ÙØ¯ÙØ© (SLA) Ø¥ÙØ§ عÙÙ Ø§ÙØ®Ø·Ø· Ø§ÙØªÙ ØªØ°ÙØ±Ùا ØµØ±Ø§ØØ©Ù.`},account:{title:"4. ØØ³Ø§Ø¨ اÙÙ Ø³ØªØ®Ø¯Ù ÙØ§Ùأ٠اÙ",content:`ØªØªØ¹ÙØ¯ Ø¨ØªÙØ¯Ù٠٠عÙÙ٠ات دÙÙÙØ© Ø¹ÙØ¯ Ø§ÙØªØ³Ø¬ÙÙ ÙØªØØ¯ÙØ«ÙØ§ باست٠رار. 932Ø£ÙØª ÙØØ¯Ù Ø§Ù٠سؤÙÙ Ø¹Ù Ø³Ø±ÙØ© Ø¨ÙØ§Ùات Ø§ÙØ§Ø¹ØªÙ اد Ø§ÙØ®Ø§ØµØ© Ø¨ÙØ ÙÙ ÙØ§ØªÙØ APIØ ÙØ±Ù ÙØ² اÙÙØµÙÙ. ÙØ¹ØªØ¨Ø± أ٠إجراء ÙØªÙ Ù Ù ØØ³Ø§Ø¨Ù ÙØ¯ ØªÙ Ø¨ÙØ§Ø³Ø·ØªÙ. 933ÙØ¬Ø¨ عÙÙÙ Ø¥Ø¨ÙØ§Øº Infernex ÙÙØ±Ùا Ø¨Ø£Ù ÙØµÙÙ ØºÙØ± Ù ØµØ±Ø Ø¨Ù. ÙØØ¸Ø± Ù Ø´Ø§Ø±ÙØ© Ø§ÙØØ³Ø§Ø¨ بÙÙ ÙÙØ§Ùات Ù ÙÙØµÙØ© ٠تعددة.`},acceptableUse:{title:"5. Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù اÙÙ ÙØ¨ÙÙ",content:`ÙØØ¸Ø± ØªÙ Ø§Ù ÙØ§ Ø§Ø³ØªØ®Ø¯Ø§Ù Ø§ÙØ®Ø¯Ù ات ٠٠أجÙ: Ø¥ÙØ´Ø§Ø¡ Ø£Ù ÙØ´Ø± Ù ØØªÙÙ ØºÙØ± ÙØ§ÙÙÙÙØ Ø£Ù Ø¨ØºÙØ¶Ø أ٠تشÙÙØ±ÙØ Ø£Ù Ù Ø¶ÙÙØ أ٠إباØÙ ÙÙØ£Ø·ÙØ§ÙØ Ø£Ù Ø§ÙØªÙا٠ØÙÙ٠اÙÙ ÙÙÙØ© اÙÙÙØ±ÙØ©Ø Ø£Ù Ø§ÙØªØØ§Ù٠عÙ٠إجراءات Ø§ÙØ£Ù Ø§ÙØ أ٠إجراء Ø³ØØ¨ ج٠اع٠ÙÙØ¨ÙØ§ÙØ§Øª (scraping)Ø Ø£Ù Ø¨Ø±ÙØ¯ ٠زعج (spam)Ø Ø£Ù Ø§ÙØªØØ§Ù ÙÙÙØ©Ø Ø£Ù ØªØ¯Ø±ÙØ¨ ÙÙ ÙØ°Ø¬ Ù ÙØ§Ùس باستخدا٠٠خرجات ÙÙ Ø§Ø°Ø¬ÙØ§Ø أ٠إعادة Ø¨ÙØ¹ اÙÙØµÙ٠إÙÙ Ø§ÙØ®Ø¯Ù ات دÙÙ Ù ÙØ§ÙÙØ© ÙØªØ§Ø¨ÙØ© Ù Ø³Ø¨ÙØ©. 934ÙÙ ÙÙ ÙÙ Infernex تعÙÙÙ Ø£Ù ØØ³Ø§Ø¨ ÙÙØªÙÙ ÙØ°Ø§ Ø§ÙØ¨Ùد ÙÙØ±Ùا ÙØ¯Ù٠إشعار ٠سبÙ.`},plansBilling:{title:"6. Ø§ÙØ®Ø·Ø·Ø اÙÙÙØªØ±Ø© ÙØ§ÙØØµØµ",content:`Ø§ÙØ£Ø³Ø¹Ø§Ø± اÙÙ Ø¹Ø±ÙØ¶Ø© ÙØ§ ØªØ´Ù Ù Ø§ÙØ¶Ø±Ø§Ø¦Ø¨ اÙÙ Ø·Ø¨ÙØ©. ÙØªÙ ÙÙØªØ±Ø© Ø§ÙØ§Ø´ØªØ±Ø§Ùات Ù ÙØ¯Ù ÙØ§Ø Ø´ÙØ±ÙÙØ§ أ٠سÙÙÙÙØ§ ØØ³Ø¨ Ø§ÙØ®Ø·Ø© اÙÙ Ø®ØªØ§Ø±Ø©Ø Ø¹Ø¨Ø± Ù Ø²ÙØ¯ Ø§ÙØ¯Ùع Ø§ÙØ®Ø§Øµ Ø¨ÙØ§. 935تتض٠٠Ù٠خطة ØØµØ© Ù Ù Ø§ÙØ±Ù ÙØ² ÙØ§ÙÙ ÙØ²Ø§Øª اÙÙ ØØ¯Ø¯Ø© ÙÙ ØµÙØØ© Ø§ÙØªØ³Ø¹Ùر. ÙØªÙ Ø¯ÙØ¹ أ٠استÙÙØ§Ù ÙØªØ¬Ø§Ùز Ø§ÙØØµØ© Ø¨Ø§ÙØ³Ø¹Ø± Ø§ÙØ²Ø§Ø¦Ø¯ Ø§ÙØØ§ÙÙØ Ø¶Ù Ù Ø§ÙØØ¯ Ø§ÙØ£ÙØµÙ Ø§ÙØ´ÙØ±Ù Ø§ÙØ°Ù ØªØØ¯Ø¯Ù. 936ØªØ¸Ù Ø§ÙØ¹Ø±Ùض Â«Ù Ø¯Ù Ø§ÙØÙØ§Ø©Â» ØµØ§ÙØØ© طاÙ٠ا ÙØ§Ù Ø§ÙØ§Ø´ØªØ±Ø§Ù ÙØ´Ø·Ùا Ù٠دÙÙØ¹Ùا دÙ٠اÙÙØ·Ø§Ø¹. Ø¨Ø§Ø³ØªØ«ÙØ§Ø¡ ٠ا ÙÙØµ عÙÙ٠اÙÙØ§ÙÙÙØ ÙØ¥Ù اÙÙ Ø¨Ø§ÙØº اÙ٠دÙÙØ¹Ø© ØºÙØ± ÙØ§Ø¨ÙØ© ÙÙØ§Ø³ØªØ±Ø¯Ø§Ø¯ عÙ٠أساس ØªÙØ§Ø³Ø¨Ù.`},externalCosts:{title:"7. Ø§ÙØªÙاÙÙÙ Ø§ÙØ®Ø§Ø±Ø¬ÙØ© ØºÙØ± اÙÙ Ø¶Ù ÙØ©",content:`ÙØ§ ÙØ´Ù Ù Ø§ÙØ®Ø·Ø·: أس٠اء اÙÙØ·Ø§ÙØ§ØªØ Ø§ÙØ§Ø³ØªØ¶Ø§ÙØ© Ø£Ù Ø§ÙØ¨Ø±Ùد Ø§ÙØ¥ÙÙØªØ±ÙÙÙ ÙØ·Ø±Ù Ø«Ø§ÙØ«Ø ØªØ±Ø§Ø®ÙØµ ÙØ§Ø´ØªØ±Ø§Ùات اÙÙ ÙØµÙات Ø§ÙØ·Ø±Ù Ø§ÙØ«Ø§ÙØ«Ø Ù ÙØ§ØªÙØ API Ø§ÙØ®Ø§ØµØ© بÙÙ (BYOK)Ø ØªØ¬Ø§ÙØ²Ø§Øª Ø§ÙØ±Ù ÙØ²Ø Ø§ÙØ¶Ø±Ø§Ø¦Ø¨ ÙØ§ÙرسÙ٠اÙ٠صرÙÙØ©Ø ÙÙØ°ÙÙ Ø§ÙØ®Ø¯Ù ات Ø§ÙØ§Ø®ØªÙØ§Ø±ÙØ©. 937ÙØ§ ÙØ´Ù ٠ذÙÙ Ø£ÙØ¶Ùا Ø§ÙØ¹ÙÙØ¯ ÙØ§ÙØªØ±Ø§Ø®ÙØµ ÙØ§ÙØ§ÙØªØ²Ø§Ù ات اÙÙØ§ÙÙÙÙØ© Ø§ÙØ¶Ø±ÙØ±ÙØ© ÙØªØ³Ø¬ÙÙ Ø´Ø±ÙØªÙÙ ÙØªÙاÙÙÙØ§ ÙØªØ´ØºÙÙÙØ§ ÙÙ Ø¨ÙØ¯ÙÙ (Ø§ÙØ³Ø¬Ù Ø§ÙØªØ¬Ø§Ø±ÙØ Ø§ÙØªØ£Ù ÙÙØ§ØªØ Ø§ÙØ¶Ø±Ø§Ø¦Ø¨ اÙÙ ØÙÙØ©Ø اÙÙØ§Ø¦ØØ© Ø§ÙØ¹Ø§Ù Ø© ÙØÙ Ø§ÙØ© Ø§ÙØ¨ÙØ§ÙØ§Øª (GDPR)Ø Ø§ÙØªØ±Ø§Ø®Ùص اÙÙ ÙÙÙØ©Ø Ø¥ÙØ®): تظ٠٠سؤÙÙÙØªÙ٠باÙÙØ§Ù Ù. 938اÙÙ ÙØ²Ø§ÙÙØ© اÙÙÙØ§Ø¦ÙØ© ÙØ´Ø±ÙتÙ٠تعت٠د عÙÙ ØªÙØ±Ø§Ø±Ø§ØªÙÙ Ø ÙØ¹Ø¯Ø¯ Ù ÙØ¸ÙÙ Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·Ùاع٠اÙÙØ´Ø·ÙÙØ ÙØØ¬Ù Ø§Ø³ØªØ®Ø¯Ø§Ù ÙÙ .`},ip:{title:"8. اÙÙ ÙÙÙØ© اÙÙÙØ±ÙØ©",content:`ØªØØªÙظ Infernex Ø¨Ø¬Ù ÙØ¹ Ø§ÙØÙÙ٠اÙ٠تعÙÙØ© باÙÙ ÙØµØ© ÙØ¨Ø±Ø§Ù Ø¬ÙØ§ ÙÙÙ Ø§Ø°Ø¬ÙØ§ ÙÙØ§Ø¬ÙØ§ØªÙØ§ ÙØ¹ÙØ§Ù Ø§ØªÙØ§ Ø§ÙØªØ¬Ø§Ø±ÙØ© ÙÙØ«Ø§Ø¦ÙÙØ§. 939ØªØØªÙظÙ٠ب٠ÙÙÙØ© Ø§ÙØ¨ÙØ§ÙØ§Øª Ø§ÙØªÙ ØªÙØ¯Ù ÙÙÙØ§. ÙÙ ØØ¯Ùد ٠ا ÙØ³Ù Ø Ø¨Ù Ø§ÙÙØ§ÙÙÙØ ت٠تÙÙÙÙ ØÙÙÙ Ø§Ø³ØªØºÙØ§Ù اÙÙ ØØªÙ٠اÙÙ ÙÙØ´Ø£ Ø¨ÙØ§Ø¡Ù عÙ٠تعÙÙ٠اتÙÙ Ø Ø¨Ø´Ø±Ø· Ø§ÙØ¯Ùع اÙÙØ§Ù Ù ÙÙÙ Ø¨Ø§ÙØº اÙ٠ستØÙØ©. 940Ø£ÙØªÙ ٠سؤÙÙÙÙ Ø¹Ù Ø§ÙØªØÙ٠٠٠أ٠اÙÙ ØØªÙ٠اÙÙ ÙÙØ´Ø£ ÙØ§ ÙÙØªÙ٠أ٠ØÙÙÙ Ø·Ø±Ù Ø«Ø§ÙØ« ÙØ¨Ù Ø£Ù Ø§Ø³ØªØºÙØ§Ù تجارÙ.`},data:{title:"9. Ø§ÙØ¨ÙØ§ÙØ§Øª Ø§ÙØ´Ø®ØµÙØ© ÙØ§ÙØ³Ø±ÙØ©",content:`ÙÙØµÙ Ù Ø¹Ø§ÙØ¬Ø© Ø§ÙØ¨ÙØ§ÙØ§Øª Ø§ÙØ´Ø®ØµÙØ© ÙÙ Ø³ÙØ§Ø³Ø© Ø§ÙØ®ØµÙØµÙØ© Ø§ÙØ®Ø§ØµØ© Ø¨ÙØ§Ø Ø§ÙØªÙ تشÙÙ Ø¬Ø²Ø¡ÙØ§ ÙØ§ ÙØªØ¬Ø²Ø£ Ù Ù ÙØ°Ù Ø§ÙØ´Ø±ÙØ·. 941تع٠٠Infernex ÙÙ Ø¹Ø§ÙØ¬ ÙØ±Ø¹Ù ÙÙØ¨ÙØ§ÙØ§Øª Ø§ÙØªÙ ØªØ¹Ø§ÙØ¬ÙÙÙØ§ عبر Ø§ÙØ®Ø¯Ù ات ÙØªÙØªØ²Ù Ø¨Ø§ÙØ§Ù تثا٠ÙÙØ§Ø¦ØØ© Ø§ÙØ¹Ø§Ù Ø© ÙØÙ Ø§ÙØ© Ø§ÙØ¨ÙØ§ÙØ§Øª (GDPR) Ù٠عاÙÙØ± Ø§ÙØ£Ù ا٠اÙ٠عÙÙØ© (Ø§ÙØªØ´ÙÙØ± Ø£Ø«ÙØ§Ø¡ اÙÙÙÙØ Ø§ÙØ¹Ø²Ù ع٠طرÙ٠أ٠ا٠٠ستÙÙ Ø§ÙØµÙÙÙØ Ø§ÙØ§Ø³ØªØ¶Ø§ÙØ© Ø§ÙØ³ÙØ§Ø¯ÙØ©).`},liability:{title:"10. اÙ٠سؤÙÙÙØ©",content:`أ٠إجراء ÙØªÙ اتخاذ٠عÙÙ Infernex AI ÙÙØ¹ ØªØØª اÙ٠سؤÙÙÙØ© اÙÙØÙØ¯Ø© ÙÙØ¹Ù ÙÙ. Ø£ÙØª ÙØØ¯Ù Ø§Ù٠سؤÙ٠ع٠اÙÙØ±Ø§Ø±Ø§Øª ÙØ§ÙÙ ØØªÙÙØ§Øª ÙØ§ÙØ§ØªØµØ§ÙØ§Øª ÙØ§Ùأت٠تة Ø§ÙØªÙ ÙØªÙ Ø¥ÙØªØ§Ø¬Ùا أ٠تشغÙÙÙØ§ عبر اÙÙ ÙØµØ©Ø ÙÙØ°ÙÙ Ø¹Ù ØªÙØ§ÙÙÙØ§ ٠ع اÙÙÙØ§ÙÙ٠اÙ٠ع٠ÙÙ Ø¨ÙØ§ ÙÙ Ø¨ÙØ¯Ù. 942Ø§ÙØ¯Ø¹Ù اÙÙÙÙ Ù ØªØ§Ø 24 ساعة Ù٠اÙÙÙ٠عبر Ø§ÙØ¨Ø±Ùد Ø§ÙØ¥ÙÙØªØ±ÙÙÙ. تخÙÙ Infernex ٠سؤÙÙÙØªÙا ÙÙ ØØ§ÙØ© Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù ØºÙØ± اÙÙ ÙØ§Ø³Ø¨ Ø£Ù ØºÙØ± اÙÙØ§ÙÙÙÙ Ø£Ù ØºÙØ± اÙÙ ØªÙØ§Ù٠٠ع Ø§ÙØ®Ø¯Ù ات Ù Ù ÙØ¨Ù Ø§ÙØ¹Ù ÙÙØ ÙÙØ°ÙÙ Ø¹Ù Ø§ÙØ£Ø¶Ø±Ø§Ø± ØºÙØ± اÙ٠باشرة (ÙÙØ¯Ø§Ù Ø§ÙØ¥ÙØ±Ø§Ø¯Ø§ØªØ Ø§ÙØ¨ÙØ§ÙØ§ØªØ اÙÙØ±Øµ Ø£Ù Ø§ÙØ³Ù عة). 943ÙÙ Ø¬Ù ÙØ¹ Ø§ÙØØ§ÙØ§ØªØ ØªÙØªØµØ± اÙ٠سؤÙÙÙØ© Ø§ÙØ¥Ø¬Ù اÙÙØ© ÙÙ Infernex عÙ٠اÙÙ Ø¨ÙØº اÙ٠دÙÙØ¹ ÙØ¹ÙÙÙØ§ Ù Ù ÙØ¨Ù Ø§ÙØ¹Ù ÙÙ Ø®ÙØ§Ù Ø§ÙØ§Ø«Ù٠عشر (12) Ø´ÙØ±Ùا Ø§ÙØ£Ø®Ùرة Ø§ÙØªÙ Ø³Ø¨ÙØª Ø§ÙØØ¯Ø« اÙ٠سبب ÙÙØ¶Ø±Ø±.`},ai:{title:"11. Ø·Ø¨ÙØ¹Ø© اÙÙØªØ§Ø¦Ø¬ Ø§ÙØªÙ ÙÙØªØ¬Ùا Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙاعÙ",content:`تعت٠د Ø§ÙØ®Ø¯Ù ات عÙÙ Ù٠اذج Ø§ÙØ°Ùاء Ø§ÙØ§ØµØ·ÙØ§Ø¹Ù Ø§ÙØ§ØØªÙ اÙÙØ©. ÙØ¯ ØªØØªÙ٠اÙÙØªØ§Ø¦Ø¬ عÙÙ Ø¹Ø¯Ù Ø¯ÙØ© أ٠سÙ٠أ٠تØÙز. 944ÙØ§ ÙØ´Ù٠اÙÙ ØØªÙ٠اÙÙ ÙÙØ´Ø£ Ø¨Ø£Ù ØØ§Ù Ù Ù Ø§ÙØ£ØÙØ§Ù ÙØµÙØØ© ÙØ§ÙÙÙÙØ© Ø£Ù Ø¶Ø±ÙØ¨ÙØ© Ø£Ù Ø·Ø¨ÙØ© أ٠٠اÙÙØ© Ø£Ù ØªÙØ¸ÙÙ ÙØ©. ÙÙØ²Ù Ø§ÙØªØÙÙ Ø§ÙØ¨Ø´Ø±Ù ÙØ¨Ù Ø£Ù Ø§Ø³ØªØ®Ø¯Ø§Ù ØØ§Ø³Ù Ø£Ù ÙØ´Ø±.`},termination:{title:"12. Ø§ÙØªØ¹ÙÙÙ ÙØ§ÙØ¥ÙÙØ§Ø¡",content:`ÙÙ ÙÙÙ٠إÙÙØ§Ø¡ اشتراÙÙÙ Ù٠أ٠ÙÙØª Ù Ù ÙÙØØ© Ø§ÙØªØÙÙ Ø§ÙØ®Ø§ØµØ© بÙÙ Ø ÙØ³Ø±Ù Ø§ÙØ¥ÙÙØ§Ø¡ ÙÙ ÙÙØ§ÙØ© ÙØªØ±Ø© اÙÙÙØªØ±Ø© Ø§ÙØØ§ÙÙØ©. 945ÙØ¬Ùز ÙÙ Infernex تعÙÙ٠أ٠إÙÙØ§Ø¡ اÙÙØµÙÙ ÙÙ ØØ§ÙØ© Ø¹Ø¯Ù Ø§ÙØ¯ÙØ¹Ø Ø£Ù Ø§ÙØªÙØ§Ù ÙØ°Ù Ø§ÙØ´Ø±ÙØ·Ø Ø£Ù Ø§ÙØ§Ø³ØªØ®Ø¯Ø§Ù Ø§ÙØ°Ù ÙØ¹Ø±Ø¶ أ٠ا٠اÙÙ ÙØµØ© ÙÙØ®Ø·Ø±. 946بعد Ø§ÙØ¥ÙÙØ§Ø¡Ø ÙØªÙ Ø§ÙØ§ØØªÙاظ Ø¨Ø¨ÙØ§ÙاتÙÙ Ù٠دة 30 ÙÙÙ ÙØ§ Ø«Ù ÙØªÙ ØØ°ÙÙØ§ ÙÙØ§Ø¦ÙÙØ§Ø Ø¨Ø§Ø³ØªØ«ÙØ§Ø¡ Ø§ÙØ§Ùتزا٠اÙÙØ§ÙÙÙÙ Ø¨Ø§ÙØ§ØØªÙاظ Ø¨ÙØ§.`},law:{title:"13. اÙÙØ§ÙÙ٠اÙ٠ع٠ÙÙ Ø¨Ù ÙØ§ÙØ³ÙØ·Ø© اÙÙØ¶Ø§Ø¦ÙØ©",content:`تخضع ÙØ°Ù Ø§ÙØ´Ø±ÙØ· ÙÙÙØ§ÙÙÙ Ø§ÙØ³ÙÙØ³Ø±Ù. ÙÙ ØØ§ÙØ© ÙØ¬Ùد ÙØ²Ø§Ø¹Ø Ø³ÙØ³Ø¹Ù Ø§ÙØ£Ø·Ø±Ø§Ù ÙØ¥Ùجاد ØÙ ÙØ¯Ù. 947ÙÙ ØØ§ÙØ© Ø¹Ø¯Ù Ø§ÙØªÙص٠إÙÙ Ø§ØªÙØ§ÙØ ÙÙ ÙØ Ø§ÙØ§Ø®ØªØµØ§Øµ اÙÙØ¶Ø§Ø¦Ù Ø§ÙØØµØ±Ù ÙÙÙ ØØ§ÙÙ Ø§ÙØ¹Ø§Ø¯ÙØ© ÙÙ ÙØ§ÙتÙ٠جÙÙÙ (سÙÙØ³Ø±Ø§)Ø Ù Ø¹ ٠راعاة اÙÙÙØ§Ø¹Ø¯ Ø§ÙØ¥ÙØ²Ø§Ù ÙØ© Ø§ÙØªÙ تØÙ ٠اÙ٠ستÙÙÙÙÙ.`},contact:{title:"14. Ø§ÙØ§ØªØµØ§Ù",content:`Infernex Corp. â Ø¬ÙÙÙØ سÙÙØ³Ø±Ø§ 948ÙÙØ¯Ø¹Ù ÙØ§ÙØ£Ø³Ø¦ÙØ© Ø§ÙØªØ¹Ø§ÙØ¯ÙØ©: [email protected] 949ÙØ±Ø¯ عÙ٠٠دار 24 ساعة عبر Ø§ÙØ¨Ø±Ùد Ø§ÙØ¥ÙÙØªØ±ÙÙÙ.`}}}},zh:{cgu:{meta:{title:"ä¸è¬ä½¿ç¨æ¡æ¬¾ â Infernex AI",description:"Infernex AI å¹³å°çä¸è¬ä½¿ç¨æ¡æ¬¾ï¼æå¡è®¿é®ãè´¦æ·ã坿¥å使ç¨ã计费ã责任ãç»æ¢åéç¨æ³å¾ã"},badge:"ååæä»¶",title:"ä¸è¬ä½¿ç¨æ¡æ¬¾",subtitle:"æ¬æ¡æ¬¾è§è Infernex AI å¹³å°åå ¶æææå¡ï¼AI Workforceãæ¨ç APIãGEOãIAO Chatï¼ç访é®å使ç¨ã",lastUpdated:"æåæ´æ°ï¼",tocTitle:"ç®å½",contactCta:"对è¿äºæ¡æ¬¾æçé®ï¼è¯·åä¿¡ç»æä»¬ï¼",sections:{definitions:{title:"1. å®ä¹",content:`âInfernexâãâæä»¬âï¼Infernex Corp.ï¼ä¸å®¶æç«äºæ¥å ç¦ï¼ç士ï¼çå ¬å¸ï¼è¿è¥å¹³å° infernex.aiã 950â客æ·âãâæ¨âï¼æå¨å¹³å°ä¸æ¥æè´¦æ·çä»»ä½èªç¶äººææ³äººã 951âæå¡âï¼æ Infernex æä¾çææåè½ï¼ç¹å«æ¯ AI Workforceãå ¼å®¹ OpenAI çæ¨ç APIãAI 代çãè¿æ¥å¨ãç½ç«çæåç¸å ³ç仪表æ¿ã
952âçæå 容âï¼ææå¡æ ¹æ®æ¨çæä»¤çæçææææ¬ã代ç ãå¾åãç½ç«ãåææå³çã`},acceptance:{title:"2. æ¡æ¬¾çæ¥å",content:`å建账æ·ã订é 计åæä½¿ç¨æå¡å³è¡¨ç¤ºæ¨å®å ¨æ¥åæ¬æ¡æ¬¾ã妿æ¨ä¸æ¥åè¿äºæ¡æ¬¾ï¼æ¨å¿ é¡»ç«å³åæ¢ä½¿ç¨æå¡ã 953Infernex å¯éæ¶ä¿®æ¹è¿äºæ¡æ¬¾ãé大修æ¹å°å¨çæåè³å° 15 天éè¿çµåé®ä»¶éç¥ã卿¤æ¥æä¹åç»§ç»ä½¿ç¨æå¡å³è¡¨ç¤ºæ¥åã`},access:{title:"3. æå¡è®¿é®",content:`æå¡ä»¥ SaaS æ¨¡å¼æä¾ï¼24/7 å¯è®¿é®ï¼ä½ç»´æ¤æä½åä¸å¯æåæ åµé¤å¤ã 954Infernex å°éååçæªæ½ç¡®ä¿é«å¯ç¨æ§ï¼ä½ä¸ä¿è¯ä¸é´æææ é误è¿è¡ãååæå¡çº§å« (SLA) ä» éç¨äºæç¡®æåç计åã`},account:{title:"4. ç¨æ·è´¦æ·åå®å ¨",content:`æ¨æ¿è¯ºå¨æ³¨åæ¶æä¾å确信æ¯å¹¶ä¿ææ´æ°ã 955æ¨å¯¹æ¨çåæ®ãAPI å¯é¥å访é®ä»¤ççä¿å¯æ§è´å ¨é¨è´£ä»»ã仿¨çè´¦æ·è¿è¡ç任使ä½åè§ä¸ºç±æ¨æ§è¡ã 956æ¨å¿ é¡»ç«å³å°ä»»ä½æªç»ææç访é®éç¥ Infernexãç¦æ¢å¤ä¸ªç¬ç«å®ä½å ±äº«ä¸ä¸ªè´¦æ·ã`},acceptableUse:{title:"5. 坿¥åç使ç¨",content:`严ç¦ä½¿ç¨æå¡ï¼çææä¼ æéæ³ã仿¨ãè¯½è°¤ãæ¬ºéªæ§æå¿ç«¥è²æ å 容ï¼ä¾µç¯ç¥è¯äº§æï¼è§é¿å®å ¨æªæ½ï¼è¿è¡å¤§è§æ¨¡æåãåå¾é®ä»¶æèº«ä»½çç¨ï¼æ ¹æ®æä»¬æ¨¡åçè¾åºè®ç»ç«äºæ¨¡åï¼ææªç»äºå 书é¢åæè½¬å®æå¡è®¿é®æéã 957Infernex å¯ç«å³æåè¿åæ¤æ¡æ¬¾çä»»ä½è´¦æ·ï¼æä¸å¦è¡éç¥ã`},plansBilling:{title:"6. 计åã计费åé é¢",content:`æ¾ç¤ºä»·æ ¼ä¸å«éç¨ç¨è´¹ã订é è´¹å°æåææææå¹´éè¿æä»¬çæ¯ä»æå¡æä¾åæ¶åï¼å ·ä½åå³äºæé计åã 958æ¯ä¸ªè®¡åé½å å«ä»¤çååè½é é¢ï¼å ·ä½è¯¦æ 请åé å®ä»·é¡µé¢ãè¶ åºé é¢ç任使¶è´¹å°æå½åè¶ é¢è´¹ç计费ï¼ä¸é为æ¨è®¾å®çæ¯æéé¢ã 959âç»èº«â伿 å¨è®¢é ææä¸ä¸é´æä»è´¹çæ åµä¸æç»ææãé¤é妿æ³å¾è§å®ï¼å·²æ¯ä»ç款项æä¸ææ¯ä¾éè¿ã`},externalCosts:{title:"7. ä¸å å«çå¤é¨è´¹ç¨",content:`计åä¸ä¸å å«ï¼ååãç¬¬ä¸æ¹æç®¡æé®ä»¶æå¡ãç¬¬ä¸æ¹è¿æ¥å¨è®¸å¯å订é ãæ¨èªå·±ç API å¯é¥ (BYOK)ã令çè¶ é¢è´¹ç¨ãç¨è´¹åé¶è¡æç»è´¹ä»¥åå¯éæå¡ã 960æ¤å¤ï¼æ¨çå ¬å¸å¨å ¶æå¨å½å®¶/å°åºæ³¨åãåè§åè¿è¥æéçååã许å¯åæ³å¾ä¹å¡ï¼åä¸ç»è®°ãä¿é©ãå½å°ç¨æ¶ãGDPRãä¸ä¸è®¸å¯çï¼ä¹ä¸å å«å¨å ï¼å®ä»¬ä»ç±æ¨å ¨æè´è´£ã 961æ¨å ¬å¸çæç»é¢ç®åå³äºæ¨çè¿ä»£ãæ´»è· AI åå·¥æ°é忍ç使ç¨éã`},ip:{title:"8. ç¥è¯äº§æ",content:`Infernex ä¿çå¹³å°ãå ¶è½¯ä»¶ãæ¨¡åãçé¢ãåçåææ¡£çå ¨é¨æå©ã 962æ¨ä¿çæ¨æäº¤çæ°æ®çæææã卿³å¾å 许çèå´å ï¼å¨å ¨é¢æ¯ä»åºä»æ¬¾é¡¹çåæä¸ï¼æ¨æ¥æä»æ¨çæä»¤çæçâçæå 容âçå¼åæã 963æ¨æè´£ä»»å¨è¿è¡ä»»ä½åä¸å¼åä¹åï¼éªè¯âçæå 容âä¸ä¾µç¯ä»»ä½ç¬¬ä¸æ¹çæå©ã`},data:{title:"9. ä¸ªäººæ°æ®åä¿å¯æ§",content:`ä¸ªäººæ°æ®çå¤ç卿们çéç§æ¿çä¸è¿è¡äºæè¿°ï¼è¯¥æ¿çæ¯æ¬æ¡æ¬¾ä¸å¯æç¼ºçä¸é¨åã 964Infernex ä½ä¸ºæ¨éè¿æå¡å¤ççæ°æ®çæ°æ®å¤çè ï¼å¹¶æ¿è¯ºéµå® GDPR 以åå ¬å¸çå®å ¨æ åï¼ä¼ è¾ä¸å å¯ãè¡çº§å®å ¨é离ã主ææç®¡ï¼ã`},liability:{title:"10. 责任",content:`å¨ Infernex AI ä¸è¿è¡ç任使ä½åç±å®¢æ·å ¨æè´è´£ãæ¨å¯¹éè¿å¹³å°çææè§¦åçå³çãå 容ãéä¿¡åèªå¨å以åå®ä»¬æ¯å¦ç¬¦åæ¨æå¨å½å®¶/å°åºçéç¨æ³å¾è´å ¨é¨è´£ä»»ã 965æä¾ 24/24 å°æ¶çµåé®ä»¶ææ¯æ¯æãInfernex 对客æ·ä¸å½ãéæ³æä¸ç¬¦åæå¡ç使ç¨ä¸æ¿æ ä»»ä½è´£ä»»ï¼ä¹ä¸å¯¹é´æ¥æå®³ï¼æ¶å ¥æå¤±ãæ°æ®æå¤±ãæºä¼æå¤±æå½¢è±¡æå¤±ï¼æ¿æ 责任ã 966å¨ä»»ä½æ åµä¸ï¼Infernex çæ»è´£ä»»é½éå¶å¨å®¢æ·å¨åçäºä»¶ä¹ååäº (12) 个æå å®é æ¯ä»çéé¢ã`},ai:{title:"11. AI çæç»æçæ§è´¨",content:`æå¡åºäºæ¦çæ§äººå·¥æºè½æ¨¡åãç»æå¯è½å å«ä¸åç¡®ãéæ¼æåè§ã 967çæå 容å¨ä»»ä½æ åµä¸åä¸æææ³å¾ãç¨å¡ãå»çãè´¢å¡æç管建议ãå¨è¿è¡ä»»ä½å ³é®ä½¿ç¨æåå¸ä¹åï¼éè¦è¿è¡äººå·¥éªè¯ã`},termination:{title:"12. æååç»æ¢",content:`æ¨å¯ä»¥éæ¶ä»ä»ªè¡¨æ¿ç»æ¢è®¢é ï¼ç»æ¢å¨å½åè®¡è´¹å¨æç»ææ¶çæã 968妿åçæªä»æ¬¾ãè¿åæ¬æ¡æ¬¾æä½¿ç¨å±åå¹³å°å®å ¨çæ åµï¼Infernex å¯ä»¥æåæç»æ¢è®¿é®æéã 969ç»æ¢åï¼æ¨çæ°æ®å°ä¿ç 30 天ï¼ç¶åæ°¸ä¹ å é¤ï¼é¤éæ³å¾è§å®éè¦ä¿çã`},law:{title:"13. éç¨æ³å¾åç®¡è¾æ",content:`æ¬æ¡æ¬¾åç士æ³å¾ç®¡è¾ãå¦åçäºè®®ï¼åæ¹åºå°½å寻æ±åå¥½è§£å³æ¹æ¡ã 970妿 æ³è¾¾æä¸è´ï¼ä¸å±ç®¡è¾æå°æäºæ¥å ç¦å·ï¼ç士ï¼çæ®éæ³é¢ï¼ä½ä¿æ¤æ¶è´¹è çå¼ºå¶æ§è§å®é¤å¤ã`},contact:{title:"14. èç³»æ¹å¼",content:`Infernex Corp. â æ¥å ç¦ï¼ç士 971æ¯æåååé®é¢ï¼[email protected] 972æä»¬ 24/24 å°æ¶éè¿çµåé®ä»¶åå¤ã`}}}},ja:{cgu:{meta:{title:"å©ç¨è¦ç´ â Infernex AI",description:"Infernex AIãã©ãããã©ã¼ã ã®å©ç¨è¦ç´ï¼ãµã¼ãã¹ã¸ã®ã¢ã¯ã»ã¹ãã¢ã«ã¦ã³ããå©ç¨è¨±è«¾ç¯å²ãè«æ±ã責任ãå¥ç´è§£é¤ãé©ç¨æ³ã"},badge:"æ³çææ¸",title:"å©ç¨è¦ç´",subtitle:"æ¬è¦ç´ã¯ãInfernex AIãã©ãããã©ã¼ã ããã³ãã®ãã¹ã¦ã®ãµã¼ãã¹ï¼AI Workforceãæ¨è«APIãGEOãIAOãã£ããï¼ã¸ã®ã¢ã¯ã»ã¹ã¨å©ç¨ãè¦å®ãã¾ãã",lastUpdated:"æçµæ´æ°æ¥ï¼",tocTitle:"ç®æ¬¡",contactCta:"æ¬è¦ç´ã«é¢ããã質åã¯ããã¡ãã¾ã§ãåãåãããã ãã",sections:{definitions:{title:"1. å®ç¾©",content:`ãInfernexãããå½ç¤¾ãï¼infernex.aiã§ã¢ã¯ã»ã¹å¯è½ãªãã©ãããã©ã¼ã ãçºè¡ãããã¸ã¥ãã¼ãï¼ã¹ã¤ã¹ï¼ã«è¨ç«ããã伿¥ãInfernex Corp.ã 973ãã¯ã©ã¤ã¢ã³ããããã客æ§ãï¼ãã©ãããã©ã¼ã ä¸ã«ã¢ã«ã¦ã³ããæã¤ãã¹ã¦ã®å人ã¾ãã¯æ³äººã 974ããµã¼ãã¹ãï¼Infernexãæä¾ãããã¹ã¦ã®æ©è½ãç¹ã«AI WorkforceãOpenAIäºææ¨è«APIãAIã¨ã¼ã¸ã§ã³ããã³ãã¯ã¿ãã¦ã§ããµã¤ãçæãé¢é£ããã·ã¥ãã¼ãã 975ãçæã³ã³ãã³ããï¼ã客æ§ã®æç¤ºã«åºã¥ãããµã¼ãã¹ã«ãã£ã¦çæãããããããããã¹ããã³ã¼ããç»åãã¦ã§ããµã¤ããåæãã¾ãã¯æ±ºå®ã`},acceptance:{title:"2. è¦ç´ã¸ã®åæ",content:`ã¢ã«ã¦ã³ãã®ä½æããã©ã³ã®è³¼èªãã¾ãã¯ãµã¼ãã¹ã®å©ç¨ã¯ãæ¬è¦ç´ã®å®å ¨ãªå諾ãæå³ãã¾ããæ¬è¦ç´ã«åæããªãå ´åã¯ãç´ã¡ã«ãµã¼ãã¹ã®ä½¿ç¨ã䏿¢ããå¿ è¦ãããã¾ãã 976Infernexã¯ãæ¬è¦ç´ããã¤ã§ã夿´ãããã¨ãã§ãã¾ããéè¦ãªå¤æ´ã¯ãæ½è¡ã®å°ãªãã¨ã15æ¥åã¾ã
976§ã«é»åã¡ã¼ã«ã§éç¥ããã¾ãããã®æ¥ä»¥éããµã¼ãã¹ãç¶ç¶ãã¦å©ç¨ããå ´åã夿´ã«åæãããã®ã¨ã¿ãªããã¾ãã`},access:{title:"3. ãµã¼ãã¹ã¸ã®ã¢ã¯ã»ã¹",content:`ãµã¼ãã¹ã¯SaaSã¢ã¼ãã§æä¾ãããã¡ã³ããã³ã¹ä½æ¥ããã³ä¸å¯æåã®å ´åãé¤ãã24æéå¹´ä¸ç¡ä¼ã§ã¢ã¯ã»ã¹å¯è½ã§ãã 977Infernexã¯ãé«ãå¯ç¨æ§ã確ä¿ããããã«åççãªææ®µãè¬ãã¾ããã䏿ã®ãªãåä½ãã¨ã©ã¼ã®ãªããã¨ãä¿è¨¼ãããã®ã§ã¯ããã¾ãããå¥ç´ä¸ã®ãµã¼ãã¹ã¬ãã«ï¼SLAï¼ã¯ãæç¤ºçã«è¨è¼ããã¦ãããã©ã³ã«ã®ã¿é©ç¨ããã¾ãã`},account:{title:"4. ã¦ã¼ã¶ã¼ã¢ã«ã¦ã³ãã¨ã»ãã¥ãªãã£",content:`ç»é²æã«ã¯æ£ç¢ºãªæ å ±ãæä¾ããå¸¸ã«ææ°ã®ç¶æ ã«ä¿ã¤ãã¨ã«åæãããã®ã¨ãã¾ãã 978ã客æ§ã¯ãã客æ§ã®è奿 å ±ãAPIãã¼ãããã³ã¢ã¯ã»ã¹ãã¼ã¯ã³ã®æ©å¯æ§ã«ã¤ãã¦åç¬ã§è²¬ä»»ãè² ãã¾ããã客æ§ã®ã¢ã«ã¦ã³ãããè¡ããããã¹ã¦ã®è¡åã¯ãã客æ§ãå®è¡ãããã®ã¨ã¿ãªããã¾ãã 979䏿£ã¢ã¯ã»ã¹ããã£ãå ´åã¯ãé æ»ãªãInfernexã«éç¥ããå¿ è¦ãããã¾ããè¤æ°ã®ç°ãªãäºæ¥ä½éã§ã®ã¢ã«ã¦ã³ãã®å ±æã¯ç¦æ¢ããã¦ãã¾ãã`},acceptableUse:{title:"5. å©ç¨å¶é",content:`ãµã¼ãã¹ã以ä¸ã®ç®çã§ä½¿ç¨ãããã¨ã¯å³ããç¦æ¢ããã¦ãã¾ãï¼éæ³ãææªçãä¸å·çã誤解ãæããã¾ãã¯å ç«¥ãã«ãçãªã³ã³ãã³ããçæã¾ãã¯æ¡æ£ãããã¨ãç¥çè²¡ç£æ¨©ã侵害ãããã¨ãã»ãã¥ãªãã£å¯¾çãåé¿ãããã¨ã大éã®ã¹ã¯ã¬ã¤ãã³ã°ãã¹ãã ãã¾ãã¯ãªããã¾ããè¡ããã¨ãå½ç¤¾ã®ã¢ãã«ã®åºåããç«¶åã¢ãã«ããã¬ã¼ãã³ã°ãããã¨ãã¾ãã¯ãäºåã®æ¸é¢ã«ããåæãªãã«ãµã¼ãã¹ã¸ã®ã¢ã¯ã»ã¹ãå販ãããã¨ã 980Infernexã¯ãæ¬æ¡é ã«éåããã¢ã«ã¦ã³ãããäºåã®éç¥ãªãã«ç´ã¡ã«åæ¢ãããã¨ãã§ãã¾ãã`},plansBilling:{title:"6. ãã©ã³ãè«æ±ãããã³ã¯ã©ã¼ã¿",content:`è¡¨ç¤ºä¾¡æ ¼ã¯ãé©ç¨ãããç¨éãé¤ãããã®ã§ãããµãã¹ã¯ãªãã·ã§ã³ã¯ã鏿ãããã©ã³ã«å¿ãã¦æåä½ã¾ãã¯å¹´åä½ã§ãå½ç¤¾ã®æ±ºæ¸ãããã¤ãã¼ãéãã¦åæãããã¾ãã 981åãã©ã³ã«ã¯ãæéãã¼ã¸ã«è©³è¿°ããã¦ãããã¼ã¯ã³ã¨æ©è½ã®ã¯ã©ã¼ã¿ãå«ã¾ãã¦ãã¾ããã¯ã©ã¼ã¿ãè¶ ããæ¶è²»ã¯ãç¾å¨æå¹ãªè¶ éæéã§è«æ±ããã¾ãããã客æ§ãè¨å®ããæé¡ä¸éå ã¨ãã¾ãã 982ãæ°¸ä¹ ããã©ã³ã¯ããµãã¹ã¯ãªãã·ã§ã³ãæå¹ã§ã䏿ãªãæ¯æããã¦ããéãæå¹ã§ããæ³ç義åãããå ´åãé¤ããæ¯æãããéé¡ã¯æ¥å²ãè¨ç®ã§ã®è¿éã¯è¡ããã¾ããã`},externalCosts:{title:"7. å«ã¾ãã¦ããªãå¤é¨è²»ç¨",content:`ãã©ã³ã«ã¯å«ã¾ãã¦ããªããã®ï¼ãã¡ã¤ã³åããµã¼ããã¼ãã£ã®ãã¹ãã£ã³ã°ã¾ãã¯ã¡ãã»ã¼ã¸ã³ã°ããµã¼ããã¼ãã£ã®ã³ãã¯ã¿ã®ã©ã¤ã»ã³ã¹ã¨ãµãã¹ã¯ãªãã·ã§ã³ãã客æ§èªèº«ã®APIãã¼ï¼BYOKï¼ããã¼ã¯ã³è¶ éæéãç¨éããã³éè¡ææ°æããªãã³ã«ãªãã·ã§ã³ãµã¼ãã¹ã 983ã¾ããã客æ§ã®å½ã§ãã¸ãã¹ãç»é²ãæºæ ãéå¶ããããã«å¿ é ã¨ãªãå¥ç´ãã©ã¤ã»ã³ã¹ãæ³ç義åï¼ååç»é²ãä¿éºãå°æ¹ç¨ãGDPRãå°éå®¶ã©ã¤ã»ã³ã¹ãªã©ï¼ãå«ã¾ãã¦ãããããããã¯ãã¹ã¦ã客æ§ã®è²¬ä»»ã¨ãªãã¾ãã 984ã客æ§ã®æçµçãªäºæ¥äºç®ã¯ãã客æ§ã®ã¤ãã¬ã¼ã·ã§ã³ãã¢ã¯ãã£ããªAI徿¥å¡ã®æ°ãããã³å©ç¨éã«ãã£ã¦ç°ãªãã¾ãã`},ip:{title:"8. ç¥çè²¡ç£æ¨©",content:`Infernexã¯ããã©ãããã©ã¼ã ããã®ã½ããã¦ã§ã¢ãã¢ãã«ãã¤ã³ã¿ã¼ãã§ã¼ã¹ã忍ãããã³ããã¥ã¡ã³ãã«å¯¾ãããã¹ã¦ã®æ¨©å©ãä¿æãã¾ãã 985ã客æ§ã¯ãã客æ§ãéä¿¡ãããã¼ã¿ã®æææ¨©ãä¿æãã¾ããæ³å¾ã§è¨±å®¹ãããç¯å²ã«ããã¦ãã客æ§ã¯ãæ¯æãæ¸ã¿ã®å ¨é¡ãæ¡ä»¶ã¨ãã¦ãã客æ§ã®æç¤ºã«åºã¥ãã¦çæãããã³ã³ãã³ãã®å©ç¨æ¨©ãä¿æãã¾ãã 986忥çã«å©ç¨ããåã«ãçæãããã³ã³ãã³ãããããªã第ä¸è ã®æ¨©å©ã侵害ãã¦ããªããã¨ã確èªãã責任ã¯ã客æ§ã«ããã¾ãã`},data:{title:"9. å人ãã¼ã¿ã¨æ©å¯ä¿æ",content:`å人ãã¼ã¿ã®å¦çã¯ãæ¬è¦ç´ã®ä¸é¨ãæ§æããå½ç¤¾ã®ãã©ã¤ãã·ã¼ããªã·ã¼ã«è¨è¼ããã¦ãã¾ãã 987Infernexã¯ãã客æ§ããµã¼ãã¹ãéãã¦å¦çãããã¼ã¿ã«ã¤ãã¦å¦çè ã¨ãã¦è¡åããGDPRããã³å ¬è¡¨ããã¦ããã»ãã¥ãªãã£åºæºï¼è»¢éä¸ã®æå·åãè¡ã¬ãã«ã»ãã¥ãªãã£ã«ããåé¢ã主権ãã¹ãã£ã³ã°ï¼ãéµå®ãããã¨ãç´æãã¾ãã`},liability:{title:"10. 責任",content:`Infernex AIä¸ã§è¡ãããããããè¡åã¯ãã¯ã©ã¤ã¢ã³ãã®åç¬ã®è²¬ä»»ã¨ãªãã¾ããã客æ§ã¯ããã©ãããã©ã¼ã ãéãã¦çæã¾ãã¯éå§ãããæ±ºå®ãã³ã³ãã³ããã³ãã¥ãã±ã¼ã·ã§ã³ãããã³èªååããªãã³ã«ã客æ§ã®å½ã®é©ç¨æ³ã¸ã®é©åæ§ã«ã¤ãã¦åç¬ã§è²¬ä»»ãè² ãã¾ãã 988æè¡ãµãã¼ãã¯ãé»åã¡ã¼ã«ã§24æéå©ç¨å¯è½ã§ããInfernexã¯ãã¯ã©ã¤ã¢ã³ãã«ãããµã¼ãã¹ã®ä¸é©åãéæ³ãã¾ãã¯ä¸é©åã
988ªä½¿ç¨ã®å ´åããªãã³ã«éæ¥çãªæå®³ï¼å£²ä¸ããã¼ã¿ãæ©ä¼ãã¾ãã¯ã¤ã¡ã¼ã¸ã®åªå¤±ï¼ã«ã¤ãã¦ä¸åã®è²¬ä»»ãè² ãã¾ããã 989ãããªãå ´åã«ããã¦ããInfernexã®å ¨ä½çãªè²¬ä»»ã¯ãçºçåå ã¨ãªãåºæ¥äºã®åã®éå»12ã¶æéã«ã客æ§ãå®éã«æ¯æã£ãéé¡ã«éå®ããã¾ãã`},ai:{title:"11. AIã«ãã£ã¦çæãããçµæã®æ§è³ª",content:`ãµã¼ãã¹ã¯ã確ççAIã¢ãã«ã«åºã¥ãã¦ãã¾ããçµæã«ã¯ä¸æ£ç¢ºããçç¥ãã¾ãã¯ãã¤ã¢ã¹ãå«ã¾ããå¯è½æ§ãããã¾ãã 990çæãããã³ã³ãã³ãã¯ããããªãå ´åãæ³å¾ãç¨åãå»çã財åãã¾ãã¯è¦å¶ã«é¢ããå©è¨ãæ§æãããã®ã§ã¯ããã¾ãããæ¹å¤çãªä½¿ç¨ã¾ãã¯å ¬éã®åã«ã¯ã人éã®æ¤è¨¼ãå¿ è¦ã§ãã`},termination:{title:"12. 忢ããã³å¥ç´è§£é¤",content:`ã客æ§ã¯ãããã·ã¥ãã¼ããããã¤ã§ããµãã¹ã¯ãªãã·ã§ã³ãè§£é¤ã§ãã¾ããè§£é¤ã¯ãç¾å¨ã®è«æ±æéã®çµäºæã«æå¹ã«ãªãã¾ãã 991Infernexã¯ã䏿ããæ¬è¦ç´ã¸ã®éåãã¾ãã¯ãã©ãããã©ã¼ã ã®å®å ¨æ§ãè ãã使ç¨ã®å ´åã«ã¯ãã¢ã¯ã»ã¹ã忢ã¾ãã¯è§£é¤ãããã¨ãã§ãã¾ãã 992å¥ç´è§£é¤å¾ãã客æ§ã®ãã¼ã¿ã¯30æ¥éä¿æããããã®å¾ãæ³çä¿æç¾©åããªãéããå®å ¨ã«åé¤ããã¾ãã`},law:{title:"13. æºæ æ³ããã³ç®¡è½",content:`æ¬è¦ç´ã¯ã¹ã¤ã¹æ³ã«æºæ ãã¾ããç´äºãçºçããå ´åãå½äºè ã¯å好çãªè§£æ±ºã試ã¿ããã®ã¨ãã¾ãã 993åæã«è³ããªãå ´åãæ¶è²»è ãä¿è·ããå¼·è¡è¦å®ãé©ç¨ãããå ´åãé¤ããã¸ã¥ãã¼ãå·ï¼ã¹ã¤ã¹ï¼ã®é常è£å¤æã«å°å±ç®¡è½æ¨©ãä¸ãããã¾ãã`},contact:{title:"14. é£çµ¡å ",content:`Infernex Corp. â ã¸ã¥ãã¼ããã¹ã¤ã¹ 994ãµãã¼ãããã³å¥ç´ã«é¢ããã質åï¼[email protected] 995é»åã¡ã¼ã«ã«ã¦24æéä½å¶ã§å¯¾å¿ãããã¾ãã`}}}},hi:{cgu:{meta:{title:"à¤à¤ªà¤¯à¥à¤ à¤à¥ सामानà¥à¤¯ शरà¥à¤¤à¥à¤ â Infernex AI",description:"Infernex AI पà¥à¤²à¥à¤à¤«à¥à¤°à¥à¤® à¤à¥ à¤à¤ªà¤¯à¥à¤ à¤à¥ सामानà¥à¤¯ शरà¥à¤¤à¥à¤: सà¥à¤µà¤¾ तठपहà¥à¤à¤, à¤à¤¾à¤¤à¤¾, सà¥à¤µà¥à¤à¤¾à¤°à¥à¤¯ à¤à¤ªà¤¯à¥à¤, बिलिà¤à¤, दà¥à¤¯à¤¤à¤¾, समापà¥à¤¤à¤¿ à¤à¤° लाà¤à¥ à¤à¤¾à¤¨à¥à¤¨à¥¤"},badge:"ठनà¥à¤¬à¤à¤§ सà¤à¤¬à¤à¤§à¥ दसà¥à¤¤à¤¾à¤µà¥à¤à¤¼",title:"à¤à¤ªà¤¯à¥à¤ à¤à¥ सामानà¥à¤¯ शरà¥à¤¤à¥à¤",subtitle:"यॠशरà¥à¤¤à¥à¤ Infernex AI पà¥à¤²à¥à¤à¤«à¥à¤°à¥à¤® à¤à¤° à¤à¤¸à¤à¥ सà¤à¥ सà¥à¤µà¤¾à¤à¤ (AI Workforce, à¤à¤¨à¥à¤«à¥à¤°à¥à¤à¤¸ API, GEO, IAO Chat) तठपहà¥à¤à¤ à¤à¤° à¤à¤ªà¤¯à¥à¤ à¤à¥ नियà¤à¤¤à¥à¤°à¤¿à¤¤ à¤à¤°à¤¤à¥ हà¥à¤à¥¤",lastUpdated:"ठà¤à¤¤à¤¿à¤® ठपडà¥à¤:",tocTitle:"विषय-सà¥à¤à¥",contactCta:"à¤à¤¨ शरà¥à¤¤à¥à¤ à¤à¥ बारॠमà¥à¤ à¤à¥à¤ पà¥à¤°à¤¶à¥à¤¨ हà¥? हमà¥à¤ यहाठलिà¤à¥à¤",sections:{definitions:{title:"1. परिà¤à¤¾à¤·à¤¾à¤à¤",content:`« Infernex », « हम »: Infernex Corp., à¤à¤¿à¤¨à¥à¤µà¤¾ (सà¥à¤µà¤¿à¤à¥à¤à¤°à¤²à¥à¤à¤¡) मà¥à¤ सà¥à¤¥à¤¾à¤ªà¤¿à¤¤ à¤à¤à¤ªà¤¨à¥, à¤à¥ infernex.ai पर à¤à¤ªà¤²à¤¬à¥à¤§ पà¥à¤²à¥à¤à¤«à¥à¤°à¥à¤® à¤à¥ पà¥à¤°à¤à¤¾à¤¶à¤ हà¥à¥¤ 996« à¤à¥à¤°à¤¾à¤¹à¤ », « à¤à¤ª »: पà¥à¤²à¥à¤à¤«à¥à¤°à¥à¤® पर à¤à¤¾à¤¤à¤¾ रà¤à¤¨à¥ वाला à¤à¥à¤ à¤à¥ पà¥à¤°à¤¾à¤à¥à¤¤à¤¿à¤ या à¤à¤¾à¤¨à¥à¤¨à¥ वà¥à¤¯à¤à¥à¤¤à¤¿à¥¤ 997« सà¥à¤µà¤¾à¤à¤ »: Infernex दà¥à¤µà¤¾à¤°à¤¾ पà¥à¤°à¤¦à¤¾à¤¨ à¤à¥ à¤à¤ सà¤à¥ सà¥à¤µà¤¿à¤§à¤¾à¤à¤, विशà¥à¤· रà¥à¤ª सॠAI Workforce, OpenAI सà¤à¤à¤¤ à¤à¤¨à¥à¤«à¥à¤°à¥à¤à¤¸ API, AI à¤à¤à¥à¤à¤, à¤à¤¨à¥à¤à¥à¤à¤°à¥à¤¸, वà¥à¤¬à¤¸à¤¾à¤à¤ à¤à¤¨à¤°à¥à¤¶à¤¨ à¤à¤° सà¤à¤¬à¤à¤§à¤¿à¤¤ डà¥à¤¶à¤¬à¥à¤°à¥à¤¡à¥¤ 998« à¤à¤¨à¤¿à¤¤ सामà¤à¥à¤°à¥ »: à¤à¤ªà¤à¥ निरà¥à¤¦à¥à¤¶à¥à¤ सॠसà¥à¤µà¤¾à¤à¤ दà¥à¤µà¤¾à¤°à¤¾ à¤à¤¤à¥à
998¤ªà¤¾à¤¦à¤¿à¤¤ à¤à¥à¤ à¤à¥ पाठ, à¤à¥à¤¡, à¤à¤µà¤¿, वà¥à¤¬à¤¸à¤¾à¤à¤, विशà¥à¤²à¥à¤·à¤£ या निरà¥à¤£à¤¯à¥¤`},acceptance:{title:"2. शरà¥à¤¤à¥à¤ à¤à¥ सà¥à¤µà¥à¤à¥à¤¤à¤¿",content:`à¤à¤¾à¤¤à¤¾ बनाना, à¤à¤¿à¤¸à¥ यà¥à¤à¤¨à¤¾ à¤à¥ सदसà¥à¤¯à¤¤à¤¾ लà¥à¤¨à¤¾ या सà¥à¤µà¤¾à¤à¤ à¤à¤¾ à¤à¥à¤ à¤à¥ à¤à¤ªà¤¯à¥à¤ à¤à¤¨ शरà¥à¤¤à¥à¤ à¤à¥ पà¥à¤°à¥à¤£ सà¥à¤µà¥à¤à¥à¤¤à¤¿ à¤à¤¾ à¤à¤ न à¤à¤°à¤¤à¤¾ हà¥à¥¤ यदि à¤à¤ª à¤à¤¨ शरà¥à¤¤à¥à¤ सॠसहमत नहà¥à¤ हà¥à¤, तॠà¤à¤ªà¤à¥ तà¥à¤°à¤à¤¤ सà¥à¤µà¤¾à¤à¤ à¤à¤¾ à¤à¤ªà¤¯à¥à¤ बà¤à¤¦ à¤à¤° दà¥à¤¨à¤¾ à¤à¤¾à¤¹à¤¿à¤à¥¤ 999Infernex à¤à¤¨ शरà¥à¤¤à¥à¤ à¤à¥ à¤à¤¿à¤¸à¥ à¤à¥ समय सà¤à¤¶à¥à¤§à¤¿à¤¤ à¤à¤° सà¤à¤¤à¤¾ हà¥à¥¤ महतà¥à¤µà¤ªà¥à¤°à¥à¤£ परिवरà¥à¤¤à¤¨à¥à¤ à¤à¥ à¤à¤¨à¤à¥ पà¥à¤°à¤à¤¾à¤µà¥ हà¥à¤¨à¥ सॠà¤à¤® सॠà¤à¤® 15 दिन पहलॠà¤à¤®à¥à¤² दà¥à¤µà¤¾à¤°à¤¾ सà¥à¤à¤¿à¤¤ à¤à¤¿à¤¯à¤¾ à¤à¤¾à¤à¤à¤¾à¥¤ à¤à¤¸ तिथि à¤à¥ बाद सà¥à¤µà¤¾à¤à¤ à¤à¤¾ निरà¤à¤¤à¤° à¤à¤ªà¤¯à¥à¤ सà¥à¤µà¥à¤à¥à¤¤à¤¿ à¤à¤¾ à¤à¤ न à¤à¤°à¤¤à¤¾ हà¥à¥¤`},access:{title:"3. सà¥à¤µà¤¾ तठपहà¥à¤à¤",content:`सà¥à¤µà¤¾à¤à¤ SaaS मà¥à¤¡ मà¥à¤ पà¥à¤°à¤¦à¤¾à¤¨ à¤à¥ à¤à¤¾à¤¤à¥ हà¥à¤, à¤à¥ 24/7 सà¥à¤²à¤ हà¥à¤, रà¤à¤°à¤à¤¾à¤µ सà¤à¤à¤¾à¤²à¤¨ à¤à¤° ठपà¥à¤°à¤¤à¥à¤¯à¤¾à¤¶à¤¿à¤¤ परिसà¥à¤¥à¤¿à¤¤à¤¿à¤¯à¥à¤ à¤à¥ ठधà¥à¤¨à¥¤ 1000Infernex बिना à¤à¤¿à¤¸à¥ बाधा या तà¥à¤°à¥à¤à¤¿ à¤à¥ निरà¤à¤¤à¤° सà¤à¤à¤¾à¤²à¤¨ à¤à¥ à¤à¤¾à¤°à¤à¤à¥ à¤à¥ बिना, à¤à¤à¥à¤ à¤à¤ªà¤²à¤¬à¥à¤§à¤¤à¤¾ सà¥à¤¨à¤¿à¤¶à¥à¤à¤¿à¤¤ à¤à¤°à¤¨à¥ à¤à¥ लिठà¤à¤à¤¿à¤¤ पà¥à¤°à¤¯à¤¾à¤¸ à¤à¤°à¤¤à¤¾ हà¥à¥¤ ठनà¥à¤¬à¤à¤§ सà¤à¤¬à¤à¤§à¥ सà¥à¤µà¤¾ सà¥à¤¤à¤° (SLA) à¤à¥à¤µà¤² à¤à¤¨ यà¥à¤à¤¨à¤¾à¤à¤ पर लाà¤à¥ हà¥à¤¤à¥ हà¥à¤ à¤à¤¿à¤¨à¤®à¥à¤ à¤à¤¨à¤à¤¾ सà¥à¤ªà¤·à¥à¤ रà¥à¤ª सॠà¤à¤²à¥à¤²à¥à¤ à¤à¤¿à¤¯à¤¾ à¤à¤¯à¤¾ हà¥à¥¤`},account:{title:"4. à¤à¤ªà¤¯à¥à¤à¤à¤°à¥à¤¤à¤¾ à¤à¤¾à¤¤à¤¾ à¤à¤° सà¥à¤°à¤à¥à¤·à¤¾",content:`à¤à¤ª पà¤à¤à¥à¤à¤°à¤£ à¤à¥ दà¥à¤°à¤¾à¤¨ सà¤à¥à¤ à¤à¤¾à¤¨à¤à¤¾à¤°à¥ पà¥à¤°à¤¦à¤¾à¤¨ à¤à¤°à¤¨à¥ à¤à¤° à¤à¤¸à¥ ठदà¥à¤¯à¤¤à¤¿à¤¤ रà¤à¤¨à¥ à¤à¤¾ वà¤à¤¨ दà¥à¤¤à¥ हà¥à¤à¥¤ 1001à¤à¤ª ठपनॠपहà¤à¤¾à¤¨à¤à¤°à¥à¤¤à¤¾à¤à¤, API à¤à¥à¤à¤à¤¿à¤¯à¥à¤ à¤à¤° à¤à¤à¥à¤¸à¥à¤¸ à¤à¥à¤à¤¨ à¤à¥ à¤à¥à¤ªà¤¨à¥à¤¯à¤¤à¤¾ à¤à¥ लिठपà¥à¤°à¥ तरह सॠà¤à¤¿à¤®à¥à¤®à¥à¤¦à¤¾à¤° हà¥à¤à¥¤ à¤à¤ªà¤à¥ à¤à¤¾à¤¤à¥ सॠà¤à¥ à¤à¤ à¤à¥à¤ à¤à¥ à¤à¤¾à¤°à¥à¤°à¤µà¤¾à¤ à¤à¤ªà¤à¥ दà¥à¤µà¤¾à¤°à¤¾ à¤à¥ à¤à¤ मानॠà¤à¤¾à¤à¤à¥à¥¤ 1002à¤à¤ªà¤à¥ à¤à¤¿à¤¸à¥ à¤à¥ ठनधिà¤à¥à¤¤ पहà¥à¤à¤ à¤à¥ बारॠमà¥à¤ Infernex à¤à¥ तà¥à¤°à¤à¤¤ सà¥à¤à¤¿à¤¤ à¤à¤°à¤¨à¤¾ हà¥à¤à¤¾à¥¤ à¤à¤ ठलà¤-ठलठसà¤à¤¸à¥à¤¥à¤¾à¤à¤ à¤à¥ बà¥à¤ à¤à¤ à¤à¤¾à¤¤à¥ à¤à¥ साà¤à¤¾ à¤à¤°à¤¨à¤¾ पà¥à¤°à¤¤à¤¿à¤¬à¤à¤§à¤¿à¤¤ हà¥à¥¤`},acceptableUse:{title:"5. सà¥à¤µà¥à¤à¤¾à¤°à¥à¤¯ à¤à¤ªà¤¯à¥à¤",content:`निमà¥à¤¨à¤²à¤¿à¤à¤¿à¤¤ à¤à¥ लिठसà¥à¤µà¤¾à¤à¤ à¤à¤¾ à¤à¤ªà¤¯à¥à¤ à¤à¤°à¤¨à¤¾ सà¤à¥à¤¤ वरà¥à¤à¤¿à¤¤ हà¥: ठवà¥à¤§, à¤à¥à¤£à¤¿à¤¤, मानहानिà¤à¤¾à¤°à¤, à¤à¥à¤°à¤¾à¤®à¤ या बाल पà¥à¤°à¥à¤¨à¥à¤à¥à¤°à¤¾à¤«à¤¿à¤ सामà¤à¥à¤°à¥ बनाना या वितरित à¤à¤°à¤¨à¤¾; बà¥à¤¦à¥à¤§à¤¿à¤ सà¤à¤ªà¤¦à¤¾ ठधिà¤à¤¾à¤°à¥à¤ à¤à¤¾ à¤à¤²à¥à¤²à¤à¤à¤¨ à¤à¤°à¤¨à¤¾; सà¥à¤°à¤à¥à¤·à¤¾ à¤à¤ªà¤¾à¤¯à¥à¤ à¤à¥ बायपास à¤à¤°à¤¨à¤¾; बड़ॠपà¥à¤®à¤¾à¤¨à¥ पर सà¥à¤à¥à¤°à¥à¤ªà¤¿à¤à¤, सà¥à¤ªà¥à¤® या पहà¤à¤¾à¤¨ à¤à¥ à¤à¥à¤°à¥ à¤à¤°à¤¨à¤¾; हमारॠमà¥à¤¡à¤² à¤à¥ à¤à¤à¤à¤ªà¥à¤ सॠà¤à¤ पà¥à¤°à¤¤à¤¿à¤¸à¥à¤ªà¤°à¥à¤§à¥ मà¥à¤¡à¤² à¤à¥ पà¥à¤°à¤¶à¤¿à¤à¥à¤·à¤¿à¤¤ à¤à¤°à¤¨à¤¾; या बिना पà¥à¤°à¥à¤µ लिà¤à¤¿à¤¤ समà¤à¥à¤¤à¥ à¤à¥ सà¥à¤µà¤¾à¤à¤ तठपहà¥à¤à¤ à¤à¥ फिर सॠबà¥à¤à¤¨à¤¾à¥¤ 1003Infernex à¤à¤¸ à¤à¤à¤¡ à¤à¥ à¤à¤²à¥à¤²à¤à¤à¤¨ मà¥à¤ à¤à¤¿à¤¸à¥ à¤à¥ à¤à¤¾à¤¤à¥ à¤à¥ बिना सà¥à¤à¤¨à¤¾ à¤à¥ तà¥à¤°à¤à¤¤ निलà¤à¤¬à¤¿à¤¤ à¤à¤° सà¤à¤¤à¤¾ हà¥à¥¤`},plansBilling:{title:"6. यà¥à¤à¤¨à¤¾à¤à¤, बिलिà¤à¤ à¤à¤° à¤à¥à¤à¤¾",content:`पà¥à¤°à¤¦à¤°à¥à¤¶à¤¿à¤¤ à¤à¥à¤®à¤¤à¥à¤ लाà¤à¥ à¤à¤°à¥à¤ à¤à¥ à¤à¥à¤¡à¤¼à¤à¤° हà¥à¤à¥¤ सदसà¥à¤¯à¤¤à¤¾à¤à¤ ठà¤à¥à¤°à¤¿à¤® रà¥à¤ª सॠबिल à¤à¥ à¤à¤¾à¤¤à¥ हà¥à¤, मासिठया वारà¥à¤·à¤¿à¤ रà¥à¤ª सॠà¤à¥à¤¨à¥ à¤à¤ यà¥à¤à¤¨à¤¾ à¤à¥ ठनà¥à¤¸à¤¾à¤°, हमारॠà¤à¥à¤à¤¤à¤¾à¤¨ पà¥à¤°à¤¦à¤¾à¤¤à¤¾ à¤à¥ माधà¥à¤¯à¤® सà¥à¥¤
1004पà¥à¤°à¤¤à¥à¤¯à¥à¤ यà¥à¤à¤¨à¤¾ मà¥à¤ à¤à¤ à¤à¥à¤à¤¨ à¤à¥à¤à¤¾ à¤à¤° सà¥à¤µà¤¿à¤§à¤¾à¤à¤ शामिल हà¥à¤ à¤à¥ मà¥à¤²à¥à¤¯ निरà¥à¤§à¤¾à¤°à¤£ पà¥à¤·à¥à¤ पर निरà¥à¤¦à¤¿à¤·à¥à¤ हà¥à¤à¥¤ à¤à¥à¤à¤¾ सॠठधिठà¤à¤¾ à¤à¥à¤ à¤à¥ à¤à¤ªà¤à¥à¤ à¤à¤¾à¤²à¥ ठधिà¤à¤¾à¤° दर पर बिल à¤à¤¿à¤¯à¤¾ à¤à¤¾à¤à¤à¤¾, à¤à¤ªà¤à¥ दà¥à¤µà¤¾à¤°à¤¾ निरà¥à¤§à¤¾à¤°à¤¿à¤¤ मासिठसà¥à¤®à¤¾ तà¤à¥¤ 1005« à¤à¤à¥à¤µà¤¨ » à¤à¤«à¤¼à¤° तब तठवà¥à¤§ रहतॠहà¥à¤ à¤à¤¬ तठसदसà¥à¤¯à¤¤à¤¾ सà¤à¥à¤°à¤¿à¤¯ à¤à¤° बिना à¤à¤¿à¤¸à¥ रà¥à¤à¤¾à¤µà¤ à¤à¥ à¤à¥à¤à¤¤à¤¾à¤¨ à¤à¥ à¤à¤¾à¤¤à¥ हà¥à¥¤ à¤à¤¬ तठà¤à¤¾à¤¨à¥à¤¨à¥ पà¥à¤°à¤¾à¤µà¤§à¤¾à¤¨ ठनà¥à¤¯à¤¥à¤¾ न à¤à¤¹à¥à¤, à¤à¥à¤à¤¤à¤¾à¤¨ à¤à¥ à¤à¤ राशि à¤à¤¨à¥à¤ªà¤¾à¤¤à¤¿à¤ रà¥à¤ª सॠवापसॠयà¥à¤à¥à¤¯ नहà¥à¤ हà¥à¥¤`},externalCosts:{title:"7. बाहरॠलाà¤à¤¤à¥à¤ शामिल नहà¥à¤ हà¥à¤",content:`यà¥à¤à¤¨à¤¾à¤à¤ मà¥à¤ शामिल नहà¥à¤ हà¥à¤: डà¥à¤®à¥à¤¨ नाम, तà¥à¤¤à¥à¤¯-पà¤à¥à¤· हà¥à¤¸à¥à¤à¤¿à¤à¤ या मà¥à¤¸à¥à¤à¤¿à¤à¤, तà¥à¤¤à¥à¤¯-पà¤à¥à¤· à¤à¤¨à¥à¤à¥à¤à¤°à¥à¤¸ à¤à¥ लाà¤à¤¸à¥à¤à¤¸ à¤à¤° सदसà¥à¤¯à¤¤à¤¾à¤à¤, à¤à¤ªà¤à¥ ठपनॠAPI à¤à¥à¤à¤à¤¿à¤¯à¤¾à¤ (BYOK), à¤à¥à¤à¤¨ ठधिà¤à¤¾à¤°, à¤à¤° à¤à¤° बà¥à¤à¤ शà¥à¤²à¥à¤, साथ हॠवà¥à¤à¤²à¥à¤ªà¤¿à¤ सà¥à¤µà¤¾à¤à¤à¥¤ 1006à¤à¤ªà¤à¥ à¤à¤à¤ªà¤¨à¥ à¤à¥ à¤à¤ªà¤à¥ दà¥à¤¶ मà¥à¤ पà¤à¤à¥à¤à¥à¤¤, ठनà¥à¤°à¥à¤ª à¤à¤° परिà¤à¤¾à¤²à¤¨ (वाणिà¤à¥à¤¯à¤¿à¤ रà¤à¤¿à¤¸à¥à¤à¤°, बà¥à¤®à¤¾, सà¥à¤¥à¤¾à¤¨à¥à¤¯ à¤à¤°à¤¾à¤§à¤¾à¤¨, GDPR, वà¥à¤¯à¤¾à¤µà¤¸à¤¾à¤¯à¤¿à¤ लाà¤à¤¸à¥à¤à¤¸, à¤à¤¦à¤¿) बनानॠà¤à¥ लिठà¤à¤µà¤¶à¥à¤¯à¤ ठनà¥à¤¬à¤à¤§, लाà¤à¤¸à¥à¤à¤¸ à¤à¤° à¤à¤¾à¤¨à¥à¤¨à¥ दायितà¥à¤µ à¤à¥ शामिल नहà¥à¤ हà¥à¤: वॠपà¥à¤°à¥ तरह सॠà¤à¤ªà¤à¥ à¤à¤¿à¤®à¥à¤®à¥à¤¦à¤¾à¤°à¥ बनॠरहतॠहà¥à¤à¥¤ 1007à¤à¤ªà¤à¥ à¤à¤à¤ªà¤¨à¥ à¤à¤¾ ठà¤à¤¤à¤¿à¤® बà¤à¤ à¤à¤ªà¤à¥ पà¥à¤¨à¤°à¤¾à¤µà¥à¤¤à¥à¤¤à¤¿à¤¯à¥à¤, सà¤à¥à¤°à¤¿à¤¯ AI à¤à¤°à¥à¤®à¤à¤¾à¤°à¤¿à¤¯à¥à¤ à¤à¥ सà¤à¤à¥à¤¯à¤¾ à¤à¤° à¤à¤ªà¤à¥ à¤à¤ªà¤¯à¥à¤ à¤à¥ मातà¥à¤°à¤¾ पर निरà¥à¤à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤`},ip:{title:"8. बà¥à¤¦à¥à¤§à¤¿à¤ सà¤à¤ªà¤¦à¤¾",content:`Infernex पà¥à¤²à¥à¤à¤«à¥à¤°à¥à¤®, à¤à¤¸à¤à¥ सà¥à¤«à¥à¤à¤µà¥à¤¯à¤°, मà¥à¤¡à¤², à¤à¤à¤à¤°à¤«à¥à¤¸, à¤à¥à¤°à¥à¤¡à¤®à¤¾à¤°à¥à¤ à¤à¤° दसà¥à¤¤à¤¾à¤µà¥à¤à¤¼à¥à¤à¤°à¤£ पर सà¤à¥ ठधिà¤à¤¾à¤°à¥à¤ à¤à¥ बरà¤à¤°à¤¾à¤° रà¤à¤¤à¤¾ हà¥à¥¤ 1008à¤à¤ª ठपनॠदà¥à¤µà¤¾à¤°à¤¾ पà¥à¤°à¤¸à¥à¤¤à¥à¤¤ à¤à¤¿à¤ à¤à¤ डà¥à¤à¤¾ à¤à¤¾ सà¥à¤µà¤¾à¤®à¤¿à¤¤à¥à¤µ बरà¤à¤°à¤¾à¤° रà¤à¤¤à¥ हà¥à¤à¥¤ à¤à¤¾à¤¨à¥à¤¨ दà¥à¤µà¤¾à¤°à¤¾ ठनà¥à¤®à¤¤ सà¥à¤®à¤¾à¤à¤ à¤à¥ à¤à¥à¤¤à¤°, à¤à¤ª दà¥à¤¯ राशियà¥à¤ à¤à¥ पà¥à¤°à¥à¤£ à¤à¥à¤à¤¤à¤¾à¤¨ à¤à¥ ठधà¥à¤¨, à¤à¤ªà¤à¥ निरà¥à¤¦à¥à¤¶à¥à¤ सॠà¤à¤¨à¤¿à¤¤ सामà¤à¥à¤°à¥ à¤à¥ शà¥à¤·à¤£ à¤à¥ ठधिà¤à¤¾à¤°à¥à¤ à¤à¥ मालिठहà¥à¤à¥¤ 1009à¤à¤ª à¤à¤¿à¤¸à¥ à¤à¥ वà¥à¤¯à¤¾à¤µà¤¸à¤¾à¤¯à¤¿à¤ शà¥à¤·à¤£ सॠपहलॠयह सतà¥à¤¯à¤¾à¤ªà¤¿à¤¤ à¤à¤°à¤¨à¥ à¤à¥ लिठà¤à¤¿à¤®à¥à¤®à¥à¤¦à¤¾à¤° हà¥à¤ à¤à¤¿ à¤à¤¨à¤¿à¤¤ सामà¤à¥à¤°à¥ à¤à¤¿à¤¸à¥ à¤à¥ तà¥à¤¸à¤°à¥ पà¤à¥à¤· à¤à¥ ठधिà¤à¤¾à¤°à¥à¤ à¤à¤¾ à¤à¤²à¥à¤²à¤à¤à¤¨ नहà¥à¤ à¤à¤°à¤¤à¥ हà¥à¥¤`},data:{title:"9. वà¥à¤¯à¤à¥à¤¤à¤¿à¤à¤¤ डà¥à¤à¤¾ à¤à¤° à¤à¥à¤ªà¤¨à¥à¤¯à¤¤à¤¾",content:`वà¥à¤¯à¤à¥à¤¤à¤¿à¤à¤¤ डà¥à¤à¤¾ à¤à¤¾ पà¥à¤°à¤¸à¤à¤¸à¥à¤à¤°à¤£ हमारॠà¤à¥à¤ªà¤¨à¥à¤¯à¤¤à¤¾ नà¥à¤¤à¤¿ मà¥à¤ वरà¥à¤£à¤¿à¤¤ हà¥, à¤à¥ à¤à¤¨ शरà¥à¤¤à¥à¤ à¤à¤¾ à¤à¤ ठà¤à¤¿à¤¨à¥à¤¨ ठà¤à¤ हà¥à¥¤ 1010Infernex à¤à¤ªà¤à¥ दà¥à¤µà¤¾à¤°à¤¾ सà¥à¤µà¤¾à¤à¤ à¤à¥ माधà¥à¤¯à¤® सॠसà¤à¤¸à¤¾à¤§à¤¿à¤¤ à¤à¤¿à¤ à¤à¤¾à¤¨à¥ वालॠडà¥à¤à¤¾ à¤à¥ लिठà¤à¤ पà¥à¤°à¥à¤¸à¥à¤¸à¤° à¤à¥ रà¥à¤ª मà¥à¤ à¤à¤¾à¤°à¥à¤¯ à¤à¤°à¤¤à¤¾ हॠà¤à¤° GDPR à¤à¥ साथ-साथ à¤à¥à¤·à¤¿à¤¤ सà¥à¤°à¤à¥à¤·à¤¾ मानà¤à¥à¤ (परिवहन मà¥à¤ à¤à¤¨à¥à¤à¥à¤°à¤¿à¤ª
1010à¥à¤¶à¤¨, पà¤à¤à¥à¤¤à¤¿-सà¥à¤¤à¤°à¥à¤¯ सà¥à¤°à¤à¥à¤·à¤¾ दà¥à¤µà¤¾à¤°à¤¾ ठलà¤à¤¾à¤µ, सà¤à¤ªà¥à¤°à¤à¥ हà¥à¤¸à¥à¤à¤¿à¤à¤) à¤à¤¾ पालन à¤à¤°à¤¨à¥ à¤à¥ लिठपà¥à¤°à¤¤à¤¿à¤¬à¤¦à¥à¤§ हà¥à¥¤`},liability:{title:"10. दà¥à¤¯à¤¤à¤¾",content:`Infernex AI पर à¤à¥ à¤à¤ à¤à¥à¤ à¤à¥ à¤à¤¾à¤°à¥à¤°à¤µà¤¾à¤ पà¥à¤°à¥ तरह सॠà¤à¥à¤°à¤¾à¤¹à¤ à¤à¥ à¤à¤¿à¤®à¥à¤®à¥à¤¦à¤¾à¤°à¥ हà¥à¥¤ à¤à¤ª पà¥à¤²à¥à¤à¤«à¥à¤°à¥à¤® à¤à¥ माधà¥à¤¯à¤® सॠà¤à¤¤à¥à¤ªà¤¨à¥à¤¨ या à¤à¥à¤°à¤¿à¤à¤° à¤à¤¿à¤ à¤à¤ निरà¥à¤£à¤¯à¥à¤, सामà¤à¥à¤°à¥, सà¤à¤à¤¾à¤° à¤à¤° सà¥à¤µà¤à¤¾à¤²à¤¨, साथ हॠà¤à¤ªà¤à¥ दà¥à¤¶ मà¥à¤ लाà¤à¥ à¤à¤¾à¤¨à¥à¤¨à¥à¤ à¤à¥ साथ à¤à¤¨à¤à¥ ठनà¥à¤°à¥à¤ªà¤¤à¤¾ à¤à¥ लिठपà¥à¤°à¥ तरह सॠà¤à¤¿à¤®à¥à¤®à¥à¤¦à¤¾à¤° हà¥à¤à¥¤ 1011à¤à¤®à¥à¤² दà¥à¤µà¤¾à¤°à¤¾ 24 à¤à¤à¤à¥ तà¤à¤¨à¥à¤à¥ सहायता à¤à¤ªà¤²à¤¬à¥à¤§ हà¥à¥¤ Infernex à¤à¥à¤°à¤¾à¤¹à¤ दà¥à¤µà¤¾à¤°à¤¾ सà¥à¤µà¤¾à¤à¤ à¤à¥ ठनà¥à¤à¤¿à¤¤, ठवà¥à¤§ या à¤à¥à¤°-ठनà¥à¤°à¥à¤ª à¤à¤ªà¤¯à¥à¤, साथ हॠठपà¥à¤°à¤¤à¥à¤¯à¤à¥à¤· à¤à¥à¤·à¤¤à¤¿ (राà¤à¤¸à¥à¤µ, डà¥à¤à¤¾, ठवसर या à¤à¤µà¤¿ à¤à¤¾ नà¥à¤à¤¸à¤¾à¤¨) à¤à¥ लिठà¤à¤¿à¤¸à¥ à¤à¥ à¤à¤¿à¤®à¥à¤®à¥à¤¦à¤¾à¤°à¥ à¤à¥ ठसà¥à¤µà¥à¤à¤¾à¤° à¤à¤°à¤¤à¤¾ हà¥à¥¤ 1012सà¤à¥ मामलà¥à¤ मà¥à¤, Infernex à¤à¥ à¤à¥à¤² दà¥à¤¯à¤¤à¤¾ à¤à¥à¤°à¤¾à¤¹à¤ दà¥à¤µà¤¾à¤°à¤¾ à¤à¤¤à¥à¤ªà¤¨à¥à¤¨ à¤à¤à¤¨à¤¾ सॠपहलॠà¤à¥ पिà¤à¤²à¥ बारह (12) महà¥à¤¨à¥à¤ à¤à¥ दà¥à¤°à¤¾à¤¨ वासà¥à¤¤à¤µ मà¥à¤ à¤à¥à¤à¤¤à¤¾à¤¨ à¤à¥ à¤à¤ राशि तठसà¥à¤®à¤¿à¤¤ हà¥à¥¤`},ai:{title:"11. AI-à¤à¤¨à¤¿à¤¤ परिणामà¥à¤ à¤à¥ पà¥à¤°à¤à¥à¤¤à¤¿",content:`सà¥à¤µà¤¾à¤à¤ सà¤à¤à¤¾à¤µà¥à¤¯ à¤à¥à¤¤à¥à¤°à¤¿à¤® बà¥à¤¦à¥à¤§à¤¿à¤®à¤¤à¥à¤¤à¤¾ मà¥à¤¡à¤² पर à¤à¤§à¤¾à¤°à¤¿à¤¤ हà¥à¤à¥¤ परिणामà¥à¤ मà¥à¤ ठशà¥à¤¦à¥à¤§à¤¿à¤¯à¤¾à¤, à¤à¥à¤ या पà¥à¤°à¥à¤µà¤¾à¤à¥à¤°à¤¹ हॠसà¤à¤¤à¥ हà¥à¤à¥¤ 1013à¤à¤¨à¤¿à¤¤ सामà¤à¥à¤°à¥ à¤à¤¿à¤¸à¥ à¤à¥ सà¥à¤¥à¤¿à¤¤à¤¿ मà¥à¤ à¤à¤¾à¤¨à¥à¤¨à¥, à¤à¤°, à¤à¤¿à¤à¤¿à¤¤à¥à¤¸à¤¾, वितà¥à¤¤à¥à¤¯ या नियामठसलाह à¤à¤¾ à¤à¤ न नहà¥à¤ à¤à¤°à¤¤à¥ हà¥à¥¤ à¤à¤¿à¤¸à¥ à¤à¥ महतà¥à¤µà¤ªà¥à¤°à¥à¤£ à¤à¤ªà¤¯à¥à¤ या पà¥à¤°à¤à¤¾à¤¶à¤¨ सॠपहलॠमानव सतà¥à¤¯à¤¾à¤ªà¤¨ à¤à¥ à¤à¤µà¤¶à¥à¤¯à¤à¤¤à¤¾ हà¥à¤¤à¥ हà¥à¥¤`},termination:{title:"12. निलà¤à¤¬à¤¨ à¤à¤° समापà¥à¤¤à¤¿",content:`à¤à¤ª ठपनॠडà¥à¤¶à¤¬à¥à¤°à¥à¤¡ सॠà¤à¤¿à¤¸à¥ à¤à¥ समय ठपनॠसदसà¥à¤¯à¤¤à¤¾ समापà¥à¤¤ à¤à¤° सà¤à¤¤à¥ हà¥à¤; समापà¥à¤¤à¤¿ वरà¥à¤¤à¤®à¤¾à¤¨ बिलिà¤à¤ ठवधि à¤à¥ ठà¤à¤¤ मà¥à¤ पà¥à¤°à¤à¤¾à¤µà¥ हà¥à¤à¥à¥¤ 1014Infernex à¤à¥à¤à¤¤à¤¾à¤¨ न à¤à¤°à¤¨à¥, à¤à¤¨ शरà¥à¤¤à¥à¤ à¤à¥ à¤à¤²à¥à¤²à¤à¤à¤¨ या पà¥à¤²à¥à¤à¤«à¥à¤°à¥à¤® à¤à¥ सà¥à¤°à¤à¥à¤·à¤¾ à¤à¥ à¤à¤¤à¤°à¥ मà¥à¤ डालनॠवालॠà¤à¤ªà¤¯à¥à¤ à¤à¥ मामलॠमà¥à¤ पहà¥à¤à¤ à¤à¥ निलà¤à¤¬à¤¿à¤¤ या समापà¥à¤¤ à¤à¤° सà¤à¤¤à¤¾ हà¥à¥¤ 1015समापà¥à¤¤à¤¿ à¤à¥ बाद, à¤à¤ªà¤à¤¾ डà¥à¤à¤¾ 30 दिनà¥à¤ तठरà¤à¤¾ à¤à¤¾à¤¤à¤¾ हॠà¤à¤° फिर सà¥à¤¥à¤¾à¤¯à¥ रà¥à¤ª सॠहà¤à¤¾ दिया à¤à¤¾à¤¤à¤¾ हà¥, à¤à¤¬ तठà¤à¤¿ à¤à¤¾à¤¨à¥à¤¨à¥ रà¥à¤ª सॠà¤à¤¸à¤à¥ à¤à¤µà¤¶à¥à¤¯à¤à¤¤à¤¾ न हà¥à¥¤`},law:{title:"13. लाà¤à¥ à¤à¤¾à¤¨à¥à¤¨ à¤à¤° à¤à¥à¤·à¥à¤¤à¥à¤°à¤¾à¤§à¤¿à¤à¤¾à¤°",content:`यॠशरà¥à¤¤à¥à¤ सà¥à¤µà¤¿à¤¸ à¤à¤¾à¤¨à¥à¤¨ दà¥à¤µà¤¾à¤°à¤¾ शासित हà¥à¤¤à¥ हà¥à¤à¥¤ विवाद à¤à¥ सà¥à¤¥à¤¿à¤¤à¤¿ मà¥à¤, पà¤à¥à¤· à¤à¤ सà¥à¤¹à¤¾à¤°à¥à¤¦à¤ªà¥à¤°à¥à¤£ समाधान à¤à¥à¤à¤¨à¥ à¤à¤¾ पà¥à¤°à¤¯à¤¾à¤¸ à¤à¤°à¥à¤à¤à¥à¥¤ 1016à¤à¤ समà¤à¥à¤¤à¥ à¤à¥ à¤à¤®à¥ मà¥à¤, à¤à¤¿à¤¨à¥à¤µà¤¾ (सà¥à¤µà¤¿à¤à¥à¤à¤°à¤²à¥à¤à¤¡) à¤à¤¾à¤µà¤¨à¥ à¤à¥ सामानà¥à¤¯ नà¥à¤¯à¤¾à¤¯à¤¾à¤²à¤¯à¥à¤ à¤à¥ विशà¥à¤· à¤à¥à¤·à¥à¤¤à¥à¤°à¤¾à
1016¤§à¤¿à¤à¤¾à¤° दिया à¤à¤¾à¤¤à¤¾ हà¥, à¤à¥ à¤à¤ªà¤à¥à¤à¥à¤¤à¤¾à¤à¤ à¤à¥ रà¤à¥à¤·à¤¾ à¤à¤°à¤¨à¥ वालॠठनिवारà¥à¤¯ नियमà¥à¤ à¤à¥ ठधà¥à¤¨ हà¥à¥¤`},contact:{title:"14. सà¤à¤ªà¤°à¥à¤",content:`Infernex Corp. â à¤à¤¿à¤¨à¥à¤µà¤¾, सà¥à¤µà¤¿à¤à¥à¤à¤°à¤²à¥à¤à¤¡ 1017सहायता à¤à¤° ठनà¥à¤¬à¤à¤§ सà¤à¤¬à¤à¤§à¥ पà¥à¤°à¤¶à¥à¤¨: [email protected] 1018हम à¤à¤®à¥à¤² दà¥à¤µà¤¾à¤°à¤¾ 24 à¤à¤à¤à¥ à¤à¤µà¤¾à¤¬ दà¥à¤¤à¥ हà¥à¤à¥¤`}}}}},am=["FR","BE","CH","LU","MC","CA","HT","SN","CI","ML","BF","NE","GN","BJ","TG","CM","TD","CF","CG","CD","GA","MG","DJ","KM","MU","SC","RW","BI","MR","MA","DZ","TN","VU","NC","PF","LB"],om=["ES","MX","AR","CO","PE","VE","CL","EC","GT","CU","BO","DO","HN","PY","SV","NI","CR","PA","UY","PR","GQ"],rm=["DE","AT","LI"],im=["BR","PT","AO","MZ","GW","TL","CV","ST"],sm=["AE","SA","QA","KW","BH","OM","IQ","JO","SY","YE","EG","SD","LY","PS"],lm=["CN","TW","HK","MO","SG"],cm=["JP"],dm=["IN"];async function um(){try{const e=await fetch("https://ipapi.co/json/",{signal:AbortSignal.timeout(3e3)});return e.ok&&(await e.json()).country_code||null}catch(e){return console.warn("Could not detect country by IP:",e),null}}function mm(e){if(!e)return"en";const t=e.toUpperCase();return am.includes(t)?"fr":sm.includes(t)?"ar":om.includes(t)?"es":rm.includes(t)?"de":im.includes(t)?"pt":lm.includes(t)?"zh":cm.includes(t)?"ja":dm.includes(t)?"hi":"en"}const Lt=["en","fr","es","de","pt","ar","zh","ja","hi"],pm=["ko","he","it","ru","tr","id"],qg=[...Lt,...pm],Xa={fr:()=>f(()=>import("./fr-B5q3fuqM.js"),[]),es:()=>f(()=>import("./es-ddgI6dwM.js"),[]),de:()=>f(()=>import("./de-DsDSdxk2.js"),[]),pt:()=>f(()=>import("./pt-BvSnBHrS.js"),[]),ar:()=>f(()=>import("./ar-D1eadWky.js"),[]),zh:()=>f(()=>import("./zh-C8WiSx9J.js"),[]),ja:()=>f(()=>import("./ja-Bm8WEA0d.js"),[]),hi:()=>f(()=>import("./hi-CfJZFMkJ.js"),[])},ot=e=>(e==null?void 0:e.split("-")[0])||"en",eo=e=>!!e&&typeof e=="object"&&!Array.isArray(e),Re=(e,t)=>{if(!t)return e;const a={...e};for(const[o,r]of Object.entries(t)){const s=a[o];a[o]=eo(s)&&eo(r)?Re(s,r):r}return a},kn=async e=>{const t=ot(e);if(t!=="en"&&Xa[t]&&!Qe.hasResourceBundle(t,"translation"))try{const a=await Xa[t]();let o=Re(a.default,Ve[t]);o=Re(o,qr[t]),o=Re(o,Dr[t]),o=Re(o,Lr[t]),o=Re(o,Fr[t]),Qe.addResourceBundle(t,"translation",o,!0,!0)}catch{}},fm=async e=>{const t=ot(e);return await kn(t),Qe.changeLanguage(t)},hm={type:"languageDetector",async:!1,detect:()=>{var a;const e=typeof window<"u"?localStorage.getItem("i18nextLng"):null;if(e&&Lt.includes(e))return e;const t=typeof navigator<"u"?(a=navigator.language)==null?void 0:a.split("-")[0]:"en";return t&&Lt.includes(t)?t:"en"},cacheUserLanguage:e=>{typeof window<"u"&&localStorage.setItem("i18nextLng",ot(e))}},gm=["ar"],gn=e=>{const t=ot(e),a=gm.includes(t)?"rtl":"ltr";document.documentElement.setAttribute("dir",a),document.documentElement.setAttribute("lang",t)},bm=()=>{var a;if(typeof window>"u")return"en";const e=localStorage.getItem("i18nextLng");if(e&&Lt.includes(e))return e;const t=(a=navigator.language)==null?void 0:a.split("-")[0];return t&&Lt.includes(t)?t:"en"},ym=bm();let ze=null;const vm=new Promise(e=>{ze=e});Qe.use(hm).use(qi).init({resources:{en:{translation:Re(Re(Re(Re(Re(Ju,Ve.en),qr.en),Dr.en),Lr.en),Fr.en)}},lng:ym,fallbackLng:"en",interpolation:{escapeValue:!1},react:{bindI18n:"languageChanged loaded",bindI18nStore:"added removed"}}).then(async()=>{try{const e=ot(Qe.language);e!=="en"&&(await kn(e),await Qe.changeLanguage(e)),gn(e)}catch{gn("en")}finally{ze==null||ze()}}).catch(()=>{gn("en"),ze==null||ze()});if(typeof window<"u"){const e=async()=>{if(!localStorage.getItem("i18nextLng"))try{const o=await um(),r=mm(o),s=ot(r);s!==ot(Qe.language)&&(await kn(s),await Qe.changeLanguage(s),localStorage.setItem("i18nextLng",s))}catch{}};localStorage.getItem("i18nextLng")||e()}globalThis.setTimeout(()=>ze==null?void 0:ze(),800);Qe.on("languageChanged",e=>{const t=ot(e);kn(t),gn(t)});function Im({children:e}){const[t,a]=i.useState(!1);return i.useEffect(()=>{let o=!1;return vm.then(()=>{o||a(!0)}),()=>{o=!0}},[]),t?n.jsx(n.Fragment,{children:e}):n.jsx("div",{className:"flex h-screen w-screen items-center justify-center bg-background",children:n.jsx("div",{className:"h-10 w-10 animate-spin rounded-full border-4 border-primary border-t-transparent"})})}function Am(){var b,x;if(typeof window>"u")return;const e=location.hostname,a=["lovableproject.com","lovableproject-dev.com","lovable.app","gpt-eng.com","gptengineer.run"].some(I=>e===I||e.endsWith("."+I)),o="[0-9a-f]{8}-[0-9a-f]{4}-4[0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}",r=a?((b=e.match(new RegExp("^(?:id-preview(?:-[a-z0-9]+)?|project)--("+o+")(?:-dev)?(?=\\.|$)","i")))==null?void 0:b[1])??((x=e.match(new RegExp("^("+o+")(?=[.-])","i")))==null?void 0:x[1]):void 0,s=window.parent&&window.parent!==window;if(!r||!s)return localStorage;const d=e.endsWith(".lovableproject-dev.com")||e.endsWith(".gpt-eng.com"),l=d?/^https:\/\/([a-z0-9-]+\.)*(lovable\.dev|gptengineer\.app)$|^http:\/\/localhost:3000$/:/^https:\/\/([a-z0-9-]+\.)*(lovable\.dev|gptengineer\.app)$/,m=location.ancestorOrigins&&location.ancestorOrigins[0]||(document.referrer?new URL(document.referrer).origin:""),u=m&&l.test(m)?[m]:d?["https://lovable.dev","http://localhost:3000"]:["https://lovable.dev"],v="lovable-preview-auth:result",h=2e3,g=()=>Math.random().toString(36).slice(2)+Date.now().toString(36),c=(I,k,P)=>new Promise(T=>{const E=g();let j=!1,D;const q=S=>{j||(j=!0,clearTimeout(D),window.removeEventListener("message",B),T(S))},B=S=>{if(u.indexOf(S.origin)<0)return;const M=S.data;M&&M.type===v&&M.requestId===E&&q(M)};window.addEventListener("message",B);const G={type:I,requestId:E,projectId:r,key:k};P!==void 0&&(G.value=P);for(const S of u)window.parent.postMessage(G,S);D=setTimeout(()=>q(null),h)});let w=!0;const y=250;return{getItem:async I=>{let k=await c("lovable-preview-auth:get",I);return!k&&w&&(await new Promise(P=>setTimeout(P,y)),k=await c("lovable-preview-auth:get",I)),w=!1,k&&k.ok&&typeof k.value=="string"?k.value===""?(localStorage.removeItem(I),null):k.value:localStorage.getItem(I)},setItem:(I,k)=>(localStorage.setItem(I,k),c("lovable-preview-auth:set",I,k).then(()=>{})),removeItem:I=>(localStorage.removeItem(I),c("lovable-preview-auth:remove",I).then(()=>{}))}}const xm="https://ikhnpszzlgazzfhvkfsn.supabase.co",wm="eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJzdXBhYmFzZSIsInJlZiI6ImlraG5wc3p6bGdhenpmaHZrZnNuIiwicm9sZSI6ImFub24iLCJpYXQiOjE3Njk1OTIwNDUsImV4cCI6MjA4NTE2ODA0NX0.2X8IcqhmBx_el3M5yiBuRrcdbzHNZm_qFq28QSywO5I",ye=Di(xm,wm,{auth:{storage:Am(),persistSession:!0,autoRefreshToken:!0}}),Dg=Object.freeze(Object.defineProperty({__proto__:null,supabase:ye},Symbol.toStringTag,{value:"Module"})),xt="hsl(var(--primary))",Sm="rgba(0, 255, 65, 0.15)",km="hsl(0, 0%, 7%)",ln="hsl(0, 0%, 14%)",jn=[{key:"photonicVerde",label:"Photonic Verde",color:"hsl(var(--primary))"},{key:"quantum",label:"Quantum Q72",color:"hsl(var(--primary))"},{key:"tsuPbit",label:"Tsu PBit",color:"hsl(var(--primary))"},{key:"mirofish",label:"MiroFish",color:"hsl(var(--primary))"}
1018],Cm=()=>{var v;const{t:e,i18n:t}=Me(),a=To(),[o,r]=i.useState(""),[s,d]=i.useState(!1),[l,m]=i.useState(null),u=async()=>{const h=o.trim();if(!h||!/^https?:\/\//i.test(h)&&!/^[a-z0-9-]+(?:\.[a-z0-9-]+)+/i.test(h)){alert(e("audit.invalidUrl"));return}let g=h;!g.startsWith("http://")&&!g.startsWith("https://")&&(g=`https://${g}`),d(!0);for(const c of jn)m(c.key),await new Promise(w=>setTimeout(w,1200));try{const{data:c,error:w}=await ye.functions.invoke("audit-quad-fusion",{body:{siteUrl:g,language:t.language}});if(w||!(c!=null&&c.auditId)){alert(e("audit.errorGeneric")),d(!1),m(null);return}sessionStorage.setItem(`audit_${c.auditId}`,JSON.stringify(c)),a(`/audit-result?id=${c.auditId}`)}catch{alert(e("audit.errorGeneric"))}finally{d(!1),m(null)}};return n.jsxs("section",{className:"px-3 py-14 sm:px-4 md:px-8 md:py-20",children:[n.jsx("h2",{className:"mx-auto max-w-3xl text-balance text-center font-mono text-lg font-bold leading-tight md:text-2xl lg:text-3xl",style:{color:"#FFFFFF"},children:e("audit.sectionTitle")}),n.jsx("p",{className:"mx-auto mt-3 max-w-2xl text-center font-mono text-[11px] leading-relaxed md:mt-4 md:text-sm",style:{color:"#FFFFFF"},children:e("audit.sectionSubtitle")}),n.jsxs("div",{className:"neon-running-border mx-auto mt-8 max-w-2xl overflow-hidden rounded-lg border md:mt-12",style:{borderColor:ln,background:km},children:[n.jsxs("div",{className:"flex items-center gap-2 border-b px-3 py-2 md:px-4 md:py-2.5",style:{borderColor:ln},children:[n.jsx("span",{className:"h-2.5 w-2.5 rounded-full md:h-3 md:w-3",style:{background:"#FF5F56"}}),n.jsx("span",{className:"h-2.5 w-2.5 rounded-full md:h-3 md:w-3",style:{background:"#FFBD2E"}}),n.jsx("span",{className:"h-2.5 w-2.5 rounded-full md:h-3 md:w-3",style:{background:"#27C93F"}}),n.jsx("span",{className:"ml-2 font-mono text-[10px] md:ml-3 md:text-xs",style:{color:xt},children:"infernex-audit-terminal"}),s&&n.jsx("span",{className:"ml-auto font-mono text-[10px] md:text-xs",style:{color:xt,animation:"terminal-blink 1s steps(2) infinite"},children:"â ANALYSE EN COURS"})]}),s?n.jsxs("div",{className:"flex flex-col items-center gap-4 p-6 md:p-10",children:[n.jsx("p",{className:"font-mono text-xs font-bold uppercase tracking-widest",style:{color:xt},children:e("audit.analyzing")}),n.jsx("div",{className:"grid w-full max-w-md grid-cols-2 gap-3",children:jn.map(h=>n.jsxs("div",{className:"flex items-center gap-2 rounded border p-2.5 font-mono text-[10px] transition-all md:text-xs",style:{borderColor:l===h.key?h.color:ln,background:l===h.key?`${h.color}10`:"transparent",color:l===h.key?h.color:"#FFFFFF"},children:[n.jsx("span",{className:"h-2 w-2 rounded-full",style:{background:l===h.key?h.color:"hsl(0,0%,25%)",boxShadow:l===h.key?`0 0 8px ${h.color}`:"none"}}),h.label,l===h.key&&n.jsx("span",{className:"ml-auto animate-pulse",children:"â"})]},h.key))}),n.jsxs("p",{className:"font-mono text-[10px] md:text-xs",style:{color:"#FFFFFF"},children:[e("audit.scanning")," ",((v=jn.find(h=>h.key===l))==null?void 0:v.label)||"..."]})]}):n.jsxs("div",{className:"flex flex-col gap-3 p-4 sm:flex-row sm:items-center md:p-6",children:[n.jsx("input",{type:"text",inputMode:"url",value:o,onChange:h=>r(h.target.value),placeholder:e("audit.placeholder"),className:"flex-1 rounded border bg-transparent px-3 py-2.5 font-mono text-xs focus:outline-none focus:ring-1 md:text-sm",style:{borderColor:ln,color:"#FFFFFF","--tw-ring-color":xt},onKeyDown:h=>h.key==="Enter"&&u()}),n.jsx("button",{onClick:u,className:"whitespace-nowrap rounded border px-4 py-2.5 font-mono text-xs font-bold uppercase tracking-wider transition-all hover:shadow-lg md:text-sm",style:{borderColor:xt,color:xt,boxShadow:`0 0 20px ${Sm}`},children:e("audit.launchButton")})]})]})]})},de={title:"Hire AI employees and create your company in one click From an idea to a future Fortune 500 company.",description:"Hire AI employees and create your company in one click From an idea to a future Fortune 500 company.",image:"https://www.infernex.ai/og-image.png",siteUrl:"https://www.infernex.ai"},Pm=({title:e,description:t=de.description,path:a="",image:o=de.image,type:r="website",noIndex:s=!1,faqItems:d,keywords:l=[],articlePublishedTime:m,articleModifiedTime:u,articleAuthor:v="Infernex AI",articleSection:h,articleTags:g})=>{const{t:c,i18n:w}=Me(),y=w.language,b=e?`${e} | Infernex AI`:de.title,x=`${de.siteUrl}${a}`,I={en:["AI inference","LLM","AI optimization","AI savings","Infernex","tokens","machine learning","OpenAI alternative","cheap LLM inference","reduce AI costs","GPT alternative","low cost AI API","LLM cost optimization","AI infrastru
1018cture","serverless AI","AI API pricing","token optimization","Claude alternative","Groq alternative","Together AI alternative","affordable AI inference","enterprise AI","AI cost reduction","LLM deployment","AI spending","GPU inference","Quad-Fusion AI","Quantum AI","Photonic AI","MiroFish","Tsu PBit","Quantum Q72","Photonic Verde","thermodynamic AI","multimodal AI","next-gen AI","deep tech","AI investment","AI developers","API IA gratuite","startup AI","AI latency","AI precision","AI energy efficiency","distributed AI agents","photonic inference","quantum computing AI","QAOA","QRNG","bio-informatique IA","protein folding AI","stochastic Langevin model"],fr:["inférence AI","LLM","optimisation IA","économies AI","Infernex","tokens","machine learning","alternative OpenAI","inférence LLM pas cher","réduire coûts IA","alternative GPT","API IA économique","optimisation coûts LLM","infrastructure IA","IA serverless","tarifs API IA","optimisation tokens","alternative Claude","alternative Groq","inférence IA abordable","IA entreprise","réduction coûts IA","déploiement LLM","dépenses IA","inférence GPU"],es:["inferencia AI","LLM","optimización IA","ahorro AI","Infernex","tokens","aprendizaje automático","alternativa OpenAI","inferencia LLM barata","reducir costos IA","alternativa GPT","API IA económica","optimización costos LLM","infraestructura IA","IA serverless","precios API IA","optimización tokens","alternativa Claude","alternativa Groq","inferencia IA asequible","IA empresarial","reducción costos IA","despliegue LLM","gastos IA","inferencia GPU"],de:["KI-Inferenz","LLM","KI-Optimierung","KI-Einsparungen","Infernex","Tokens","maschinelles Lernen","OpenAI Alternative","günstige LLM-Inferenz","KI-Kosten senken","GPT Alternative","kostengünstige KI-API","LLM-Kostenoptimierung","KI-Infrastruktur","serverlose KI","KI-API-Preise","Token-Optimierung","Claude Alternative","Groq Alternative","erschwingliche KI-Inferenz","Unternehmens-KI","KI-Kostenreduzierung","LLM-Bereitstellung","KI-Ausgaben","GPU-Inferenz"],pt:["inferência AI","LLM","otimização IA","economia AI","Infernex","tokens","aprendizado de máquina","alternativa OpenAI","inferência LLM barata","reduzir custos IA","alternativa GPT","API IA econômica","otimização custos LLM","infraestrutura IA","IA serverless","preços API IA","otimização tokens","alternativa Claude","alternativa Groq","inferência IA acessÃvel","IA empresarial","redução custos IA","implantação LLM","gastos IA","inferência GPU"]},k=I[y]||I.en,P=[...new Set([...l,...k])].join(", "),T={en:"en_US",fr:"fr_CH",es:"es_ES",de:"de_DE",pt:"pt_BR",ar:"ar_AE",hi:"hi_IN",ja:"ja_JP",zh:"zh_CN"},E=T[y]||"en_US",j=Object.entries(T).filter(([M])=>M!==y).map(([,M])=>M),D={"@context":"https://schema.org","@type":"SoftwareApplication",name:"Infernex AI",applicationCategory:"BusinessApplication",operatingSystem:"Web",description:de.description,url:de.siteUrl,offers:{"@type":"AggregateOffer",priceCurrency:"CHF",lowPrice:"0",highPrice:"199",offerCount:"3"},aggregateRating:{"@type":"AggregateRating",ratingValue:"4.9",reviewCount:"127"},provider:{"@type":"Organization",name:"Infernex",url:de.siteUrl,logo:`${de.siteUrl}/og-image.png`,address:{"@type":"PostalAddress",addressLocality:"Lausanne",addressCountry:"CH"}}},q={"@context":"https://schema.org","@type":"Organization",name:"Infernex",url:de.siteUrl,logo:`${de.siteUrl}/og-image.png`,sameAs:["https://twitter.com/infernex_ai","https://linkedin.com/company/infernex","https://github.com/infernex"],contactPoint:{"@type":"ContactPoint",email:"[email protected]",contactType:"customer service"}},B=d&&d.length>0?{"@context":"https://schema.org","@type":"FAQPage",mainEntity:d.map(M=>({"@type":"Question",name:M.question,acceptedAnswer:{"@type":"Answer",text:M.answer}}))}:null,G=r==="article"?{"@context":"https://schema.org","@type":"BlogPosting",headline:(e||b||"").slice(0,110),description:t,image:o,url:x,datePublished:m||new Date().toISOString(),dateModified:u||m||new Date().toISOString(),author:{"@type":"Organization",name:v,url:de.siteUrl},publisher:{"@type":"Organization",name:"Infernex AI",logo:{"@type":"ImageObject",url:`${de.siteUrl}/og-image.png`}},mainEntityOfPage:{"@type":"WebPage","@id":x},...h?{articleSection:h}:{},...g?{keywords:g.join(", ")}:{},inLanguage:y}:null,S=r==="article"?{"@context":"https://schema.org","@type":"BreadcrumbList",itemListElement:[{"@type":"ListItem",position:1,name:"Blog",item:`${de.siteUrl}/blog`},{"@type":"ListItem",position:2,name:e||b,item:x}]}:null;return n.jsxs(Nc,{children:[n.jsx("title",{children:b}),n.jsx("meta",{name:"title",content:b}),n.jsx("meta",{name:"description",content:t}),n.jsx("meta",{name:"keywords",content:P}),n.jsx("html",{lang:y}),Lt.map(M=>n.jsx("link",{rel:"alternate",hrefLang:M,href:`${de.siteUrl}${a}?lang=${M}`},M)),n.jsx("link",{rel:"alternate",hrefLang:"x-default",href:`${de.siteUrl}${a}`}),s?n.jsx("meta",{name:"robots",content:"noindex, nofollow"}):n.jsx("meta",{name:"robots",content:"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1"}),n.jsx("link",{rel:"canonical",href:x}),n.jsx("meta",{property:"og:type",content:r}),n.jsx("meta",{property:"og:url",content:x}),n.jsx("meta",{property:"og:title",content:b}),n.jsx("meta",{property:"og:description",content:t}),n.jsx("meta",{property:"og:image",content:o}),n.jsx("meta",{property:"og:site_name",content:"Infernex AI"}),n.jsx("meta",{property:"og:locale",content:E}),j.map(M=>n.jsx("meta",{property:"og:locale:alternate",content:M},M)),r==="article"&&m&&n.jsx("meta",{property:"article:published_time",content:m}),r==="article"&&u&&n.jsx("meta",{property:"article:modified_time",content:u}),r==="article"&&v&&n.jsx("meta",{property:"article:author",content:v}),r==="article"&&h&&n.jsx("meta",{property:"article:section",content:h}
1018),r==="article"&&(g==null?void 0:g.map(M=>n.jsx("meta",{property:"article:tag",content:M},M))),n.jsx("meta",{name:"twitter:card",content:"summary_large_image"}),n.jsx("meta",{name:"twitter:url",content:x}),n.jsx("meta",{name:"twitter:title",content:b}),n.jsx("meta",{name:"twitter:description",content:t}),n.jsx("meta",{name:"twitter:image",content:o}),n.jsx("meta",{name:"twitter:site",content:"@infernex_ai"}),n.jsx("meta",{name:"twitter:creator",content:"@infernex_ai"}),n.jsx("script",{type:"application/ld+json",children:JSON.stringify(D)}),n.jsx("script",{type:"application/ld+json",children:JSON.stringify(q)}),G&&n.jsx("script",{type:"application/ld+json",children:JSON.stringify(G)}),B&&n.jsx("script",{type:"application/ld+json",children:JSON.stringify(B)}),S&&n.jsx("script",{type:"application/ld+json",children:JSON.stringify(S)})]})},Ht=i.forwardRef(({className:e,type:t,...a},o)=>n.jsx("input",{type:t,className:se("flex h-10 w-full rounded-md border border-input bg-background px-3 py-2 text-base ring-offset-background file:border-0 file:bg-transparent file:text-sm file:font-medium file:text-foreground placeholder:text-muted-foreground focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-ring focus-visible:ring-offset-2 disabled:cursor-not-allowed disabled:opacity-50 md:text-sm",e),ref:o,...a}));Ht.displayName="Input";const oe="hsl(var(--primary))",to="rgba(0,255,65,0.15)",no="hsl(var(--background))",Gn="#111111",Bn="#242424",Tm={MMLU:us,"ARC-AGI-2":vs,Latency:ma,DecodingTrust:ir,Throughput:ds,"Energy Efficiency":rr},Em=()=>{const{t:e}=Me(),[t,a]=i.useState([]),[o,r]=i.useState([]),[s,d]=i.useState(!0),[l,m]=i.useState(null);i.useEffect(()=>{(async()=>{const[{data:w},{data:y}]=await Promise.all([ye.from("benchmark_results").select("metric_name, score_normalized, proof_hash, metric_category").eq("model_name","INFERNEX").order("created_at",{ascending:!1}),ye.from("competitor_cache").select("competitor_name, metric_name, score_normalized")]);if(w){const b=new Set,x=w.filter(E=>b.has(E.metric_name)?!1:(b.add(E.metric_name),!0));a(x);const I=[...x].sort((E,j)=>E.metric_name.localeCompare(j.metric_name)).map(E=>`${E.metric_name}:${E.score_normalized}`).join("|"),k=new TextEncoder,P=await crypto.subtle.digest("SHA-256",k.encode(I)),T=Array.from(new Uint8Array(P));m(T.map(E=>E.toString(16).padStart(2,"0")).join(""))}y&&r(y),d(!1)})()},[]);const u=i.useMemo(()=>t.map(w=>w.metric_name).map(w=>{var x;const y={metric:e(`benchmarkWidget.metrics.${w}`,w)};return y.INFERNEX=((x=t.find(I=>I.metric_name===w))==null?void 0:x.score_normalized)??0,[...new Set(o.map(I=>I.competitor_name))].forEach(I=>{var k;y[I]=((k=o.find(P=>P.competitor_name===I&&P.metric_name===w))==null?void 0:k.score_normalized)??0}),y}),[t,o,e]),v=[...new Set(o.map(c=>c.competitor_name))],h={"GPT-5":"#FF6B6B","Claude 4":"#FFB347","Gemini 2.5":"#4FC3F7"},g=i.useMemo(()=>t.length?+(t.reduce((c,w)=>c+w.score_normalized,0)/t.length).toFixed(1):0,[t]);return s?n.jsx("div",{className:"flex items-center justify-center py-20",children:n.jsx("div",{className:"h-8 w-8 animate-spin rounded-full border-2 border-t-transparent",style:{borderColor:oe}})}):n.jsx("section",{className:"px-3 py-12 sm:px-4 md:px-8 md:py-16",children:n.jsxs("div",{className:"mx-auto max-w-6xl",children:[n.jsxs(ct.div,{initial:{opacity:0,y:20},whileInView:{opacity:1,y:0},viewport:{once:!0},className:"mb-6 text-center md:mb-10",children:[n.jsx("h2",{className:"font-mono text-xl font-bold uppercase tracking-widest sm:text-2xl md:text-3xl",style:{color:oe},children:e("benchmarkWidget.title")}),n.jsx("p",{className:"mt-2 font-mono text-[10px] md:text-xs",style:{color:"#FFFFFF"},children:e("benchmarkWidget.subtitle")})]}),n.jsxs("div",{className:"grid grid-cols-1 gap-4 md:grid-cols-3 md:gap-6",children:[n.jsxs(ct.div,{initial:{opacity:0,scale:.95},whileInView:{opacity:1,scale:1},viewport:{once:!0},className:"flex flex-col items-center justify-center rounded-xl border p-6 md:p-8",style:{borderColor:`${oe}33`,background:Gn},children:[n.jsxs("div",{className:"relative",children:[n.jsx("div",{className:"flex h-28 w-28 items-center justify-center rounded-full border-2 md:h-36 md:w-36",style:{borderColor:oe,boxShadow:`0 0 30px ${to}, 0 0 60px ${to}`},children:n.jsx("span",{className:"font-mono text-3xl font-bold md:text-4xl",style:{color:oe},children:g})}),n.jsx("div",{className:"absolute -bottom-1 left-1/2 -translate-x-1/2 roun
1018ded-full px-3 py-0.5 font-mono text-[9px] font-bold uppercase",style:{background:oe,color:no},children:"/100"})]}),n.jsx("p",{className:"mt-4 font-mono text-[10px] font-bold uppercase tracking-widest md:text-xs",style:{color:oe},children:e("benchmarkWidget.globalScore")}),n.jsx("p",{className:"mt-1 font-mono text-[9px] md:text-[10px]",style:{color:"#FFFFFF"},children:e("benchmarkWidget.liveData")})]}),n.jsxs(ct.div,{initial:{opacity:0,scale:.95},whileInView:{opacity:1,scale:1},viewport:{once:!0},transition:{delay:.1},className:"col-span-1 rounded-xl border p-4 md:col-span-2 md:p-6",style:{borderColor:Bn,background:Gn},children:[n.jsx("h3",{className:"mb-2 font-mono text-[10px] font-bold uppercase tracking-widest md:text-xs",style:{color:oe},children:e("benchmarkWidget.radarTitle")}),n.jsx(Pi,{width:"100%",height:280,children:n.jsxs(Ti,{data:u,cx:"50%",cy:"50%",outerRadius:"70%",children:[n.jsx(Ei,{stroke:"#333"}),n.jsx(Ri,{dataKey:"metric",tick:{fill:"#FFFFFF",fontSize:10,fontFamily:"monospace"}}),n.jsx(Oi,{angle:30,domain:[0,100],tick:!1,axisLine:!1}),n.jsx(Ea,{name:"INFERNEX",dataKey:"INFERNEX",stroke:oe,fill:oe,fillOpacity:.25,strokeWidth:2}),v.map(c=>n.jsx(Ea,{name:c,dataKey:c,stroke:h[c]||"#666",fill:h[c]||"#666",fillOpacity:.08,strokeWidth:1,strokeDasharray:"4 4"},c)),n.jsx(Mi,{wrapperStyle:{fontFamily:"monospace",fontSize:10}})]})})]})]}),n.jsx("div",{className:"mt-4 grid grid-cols-2 gap-3 sm:grid-cols-3 md:mt-6 md:grid-cols-6 md:gap-4",children:t.map((c,w)=>{const y=Tm[c.metric_name]||ma,b=o.filter(I=>I.metric_name===c.metric_name).sort((I,k)=>k.score_normalized-I.score_normalized)[0],x=b?+(c.score_normalized-b.score_normalized).toFixed(1):0;return n.jsxs(ct.div,{initial:{opacity:0,y:15},whileInView:{opacity:1,y:0},viewport:{once:!0},transition:{delay:w*.05},className:"rounded-lg border p-3 transition-colors hover:border-primary md:p-4 benchmark-card-glow",style:{borderColor:Bn,background:Gn,animationDelay:`${w*.8}s`},children:[n.jsx(y,{className:"mb-2 h-4 w-4 md:h-5 md:w-5",style:{color:oe}}),n.jsx("div",{className:"font-mono text-lg font-bold md:text-xl",style:{color:oe},children:c.score_normalized}),n.jsx("div",{className:"font-mono text-[9px] font-bold uppercase tracking-wider md:text-[10px]",style:{color:"#FFFFFF"},children:e(`benchmarkWidget.metrics.${c.metric_name}`,c.metric_name)}),x>0&&n.jsxs("div",{className:"mt-1 font-mono text-[8px] md:text-[9px]",style:{color:oe},children:["+",x," ",e("benchmarkWidget.vsLeader")]})]},c.metric_name)})}),l&&n.jsxs(ct.div,{initial:{opacity:0}
1018,whileInView:{opacity:1},viewport:{once:!0},className:"mt-4 flex flex-wrap items-center justify-center gap-2 rounded-lg border p-3 md:mt-6 md:p-4",style:{borderColor:Bn,background:no},children:[n.jsx(rr,{className:"h-3.5 w-3.5",style:{color:oe}}),n.jsxs("span",{className:"font-mono text-[9px] font-bold uppercase tracking-wider md:text-[10px]",style:{color:"#FFFFFF"},children:[e("benchmarkWidget.proofHash"),":"]}),n.jsxs("code",{className:"break-all font-mono text-[8px] md:text-[9px]",style:{color:"#FFFFFF"},children:["SHA-256:",l.slice(0,32),"..."]}),n.jsxs(Ee,{to:`/verify/benchmark/${l}`,className:"inline-flex items-center gap-1 rounded border px-2 py-1 font-mono text-[9px] font-bold uppercase tracking-wider transition-all hover:bg-primary/10 md:text-[10px]",style:{borderColor:oe,color:oe},children:["Vérifier la preuve",n.jsx(hs,{className:"h-3 w-3"})]})]})]})})},ie="#4F46E5",Rm="#FBFBFD",Un="#0B1220",Wt="#4B5563",ra="#E5E7F5",Om=[{key:"gdpr",fallback:"GDPR Compliant"},{key:"tls13",fallback:"TLS 1.3 Encryption"},{key:"swissHosted",fallback:"Swiss-Hosted Infrastructure"},{key:"rlsAuth",fallback:"Row-Level Security"}],Mm=[{key:"platform",href:"/platform",fallback:"Platform"},{key:"geo",href:"/geo",fallback:"GEO"},{key:"iaoChat",href:"/iao",fallback:"IAO Chat"}],qm=()=>{const{t:e}=Me();return n.jsxs("footer",{className:"relative",style:{background:Rm},children:[n.jsx("div",{className:"h-px w-full",style:{background:`linear-gradient(90deg, transparent 0%, ${ie}66 50%, transparent 100%)`}}),n.jsxs("div",{className:"mx-auto max-w-6xl px-4 py-8 md:px-6 md:py-12",children:[n.jsx("p",{className:"mx-auto max-w-4xl font-sans text-xs leading-relaxed md:text-sm",style:{color:"#FFFFFF",textAlign:"center"},children:n.jsxs("span",{style:{color:Un},children:[n.jsx("span",{style:{color:ie,fontWeight:700},children:"INFERNEX"})," ",e("institutionalFooter.safeHarbor")," ",n.jsx("span",{style:{color:ie,fontWeight:700},children:"INFERNEX SOVEREIGN TRUST"}),"."]})}),n.jsx("div",{className:"mt-6 flex flex-wrap justify-center gap-3 md:mt-8 md:gap-4",children:Om.map(t=>n.jsx(Dm,{label:e(`institutionalFooter.badgeLabels.${t.key}`,t.fallback),sublabel:e(`institutionalFooter.badges.${t.key}`,t.fallback)},t.key))}),n.jsx("div",{className:"mt-6 flex flex-wrap items-center justify-center gap-3 md:mt-8 md:gap-4",children:Mm.map(t=>n.jsx(Ee,{to:t.href,className:"rounded-md border px-4 py-2 text-sm font-semibold tracking-tight transition-all duration-300",style:{borderColor:ra,color:Un},onMouseEnter:a=>{a.currentTarget.style.borderColor=ie,a.currentTarget.style.color=ie},onMouseLeave:a=>{a.currentTarget.style.borderColor=ra,a.currentTarget.style.color=Un},children:e(`institutionalFooter.modules.${t.key}`,t.fallback)},t.key))}),n.jsx("div",{className:"mt-6 flex flex-wrap justify-center gap-x-4 gap-y-1 md:mt-8",children:[{key:"legalMentions",href:"/privacy"},{key:"sovereignTerms",href:"/cgu"}].map(({key:t,href:a})=>n.jsx("a",{href:a,className:"font-sans text-xs transition-all hover:underline md:text-sm",style:{color:Wt},onMouseEnter:o=>o.currentTarget.style.color=ie,onMouseLeave:o=>o.currentTarget.style.color=Wt,children:e(`institutionalFooter.links.${t}`)},t))}),n.jsxs("div",{className:"mt-6 flex items-center justify-center gap-2 md:mt-8",children:[n.jsx("span",{className:"h-2 w-2 rounded-full animate-pulse",style:{background:ie,boxShadow:`0 0 8px ${ie}, 0 0 16px ${ie}44`}}),n.jsx("span",{className:"font-sans text-xs font-semibold uppercase tracking-wide md:text-sm",style:{color:ie,textShadow:`0 0 10px ${ie}66`},children:e("institutionalFooter.networkStatus")})]}),n.jsxs("p",{className:"mt-4 text-center font-sans text-xs",style:{color:Wt},children:["© 2026 Infernex Corp. â Genève ð¨ð â ",e("institutionalFooter.allRights")]})]})]})},Dm=({label:e,sublabel:t})=>{const[a,o]=i.useState(!1);return n.jsxs("div",{className:"cursor-default rounded border px-3 py-1.5 font-sans text-xs font-semibold uppercase tracking-wide transition-all duration-300 md:px-4 md:py-2 md:text-sm",style:{borderColor:a?ie:ra,color:a?ie:Wt,background:a?`${ie}11`:"transparent",boxShadow:a?`0 0 15px ${ie}22`:"none"},onMouseEnter:()=>o(!0),onMouseLeave:()=>o(!1),title:t,children:[e,n.jsx("span",{className:"ml-1.5 text-[7px] md:text-[8px]",style:{color:a?ie:Wt},children:"â"})]})},Lm=ar("inline-flex items-center justify-center gap-2 whitespace-nowrap rounded-lg text-sm font-semibold transition-all duration-300 focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-ring focus-visible:ring-offset-2 disabled:pointer-events-none disabled:opacity-50 [&_svg]:pointer-events-none [&_svg]:size-4 [&_svg]:shrink-0",{variants:{variant:{default:"bg-primary text-primary-foreground shadow-lg hover:bg-primary/90 hover:shadow-xl hover:shadow-
1018primary/25",destructive:"bg-destructive text-destructive-foreground shadow-sm hover:bg-destructive/90",outline:"border border-border bg-transparent text-foreground hover:bg-secondary hover:border-primary/50",secondary:"bg-secondary text-secondary-foreground shadow-sm hover:bg-secondary/80",ghost:"text-foreground hover:bg-secondary hover:text-foreground",link:"text-primary underline-offset-4 hover:underline",hero:"bg-transparent text-foreground border border-primary/40 hover:border-primary/70 hover:bg-primary/5 hover:scale-[1.02] active:scale-[0.98]","hero-outline":"border border-primary/30 bg-transparent text-foreground hover:border-primary/50 hover:bg-primary/5",glow:"bg-primary text-primary-foreground shadow-lg shadow-primary/40 hover:shadow-xl hover:shadow-primary/60"},size:{default:"h-10 px-5 py-2",sm:"h-9 rounded-md px-4 text-xs",lg:"h-12 rounded-lg px-8 text-base",xl:"h-14 rounded-xl px-10 text-lg",icon:"h-10 w-10"}},defaultVariants:{variant:"default",size:"default"}}),ia=i.forwardRef(({className:e,variant:t,size:a,asChild:o=!1,...r},s)=>{const d=o?xi:"button";return n.jsx(d,{className:se(Lm({variant:t,size:a,className:e})),ref:s,...r})});ia.displayName="Button";const De="hsl(var(--primary))",Qn="hsl(var(--background))",Fm=["gmail.com","yahoo.com","hotmail.com","outlook.com","aol.com","icloud.com","mail.com","protonmail.com","zoho.com","yandex.com"],Nm=e=>{var a;const t=(a=e.split("@")[1])==null?void 0:a.toLowerCase();return!!t&&!Fm.includes(t)},_m=({isOpen:e,onClose:t})=>{const{t:a}=Me(),[o,r]=i.useState(""),[s,d]=i.useState(""),[l,m]=i.useState(""),[u,v]=i.useState(!1),[h,g]=i.useState(!1),[c,w]=i.useState(""),y=async b=>{if(b.preventDefault(),w(""),!(!o||!s||!l)){if(!Nm(o)){w(a("identityGateway.corporateOnly","Please use a corporate email address."));return}v(!0);try{const x=o.split("@")[1],{error:I}=await ye.from("institutional_leads").insert({email:o,organization_name:s,company_domain:x,consent_given:!0,consent_text:"Opt-in for institutional sovereign access â GDPR compliant",source:"identity_gateway"});if(I)throw I;g(!0),Wn({title:a("identityGateway.successTitle","Access Requested"),description:a("identityGateway.successDesc","Your institutional access request has been submitted.")})}catch{Wn({title:"Error",description:a("auto.3de10edbe2","Failed to submit. Please try again."),variant:"destructive"})}finally{v(!1)}}};return n.jsx(si,{children:e&&n.jsx(ct.div,{initial:{opacity:0}
1018,animate:{opacity:1},exit:{opacity:0},className:"fixed inset-0 z-[100] flex items-center justify-center bg-black/80 p-4 backdrop-blur-sm",onClick:t,children:n.jsxs(ct.div,{initial:{opacity:0,scale:.9,y:20},animate:{opacity:1,scale:1,y:0},exit:{opacity:0,scale:.9,y:20},className:"relative w-full max-w-md rounded-xl border p-6",style:{borderColor:`${De}44`,background:Qn,boxShadow:`0 0 40px ${De}11, 0 0 80px ${De}08`},onClick:b=>b.stopPropagation(),children:[n.jsx("button",{onClick:t,className:"absolute right-3 top-3 rounded p-1 transition-colors hover:bg-white/10",children:n.jsx(xn,{className:"h-4 w-4",style:{color:"#666"}})}),h?n.jsxs("div",{className:"flex flex-col items-center py-8 text-center",children:[n.jsx(fs,{className:"mb-4 h-12 w-12",style:{color:De}}),n.jsx("h3",{className:"font-mono text-lg font-bold",style:{color:De},children:a("identityGateway.accessGranted","Request Submitted")}),n.jsx("p",{className:"mt-2 font-mono text-xs",style:{color:"#666"},children:a("identityGateway.reviewMessage","Our sovereign access team will review your request within 24h.")}),n.jsx(ia,{className:"mt-6 font-mono text-xs uppercase",style:{background:De,color:Qn},onClick:t,children:a("identityGateway.close","Close")})]}):n.jsxs(n.Fragment,{children:[n.jsxs("div",{className:"mb-6 text-center",children:[n.jsx("div",{className:"mx-auto mb-3 flex h-12 w-12 items-center justify-center rounded-full border",style:{borderColor:`${De}44`},children:n.jsx(ir,{className:"h-6 w-6",style:{color:De}})}),n.jsx("h3",{className:"font-mono text-sm font-bold uppercase tracking-widest",style:{color:De},children:a("identityGateway.title","Sovereign Gateway")}),n.jsx("p",{className:"mt-1 font-mono text-[10px]",style:{color:"#555"},children:a("identityGateway.subtitle","Institutional access for verified organizations â GDPR compliant")})]}),n.jsxs("form",{onSubmit:y,className:"space-y-4",children:[n.jsxs("div",{children:[n.jsxs("label",{className:"mb-1 flex items-center gap-1.5 font-mono text-[10px] uppercase tracking-wider",style:{color:"#666"},children:[n.jsx(ys,{className:"h-3 w-3"}),a("identityGateway.contactName","Contact Name")]}),n.jsx(Ht,{value:l,onChange:b=>m(b.target.value),placeholder:a("identityGateway.placeholders.name","John Smith"),required:!0,maxLength:100,className:"border-[#333] bg-[#111] font-mono text-xs text-foreground placeholder:text-[#444]",style:{borderColor:"#333"}})]}),n.jsxs("div",{children:[n.jsxs("label",{className:"mb-1 flex items-center gap-1.5 font-mono text-[10px] uppercase tracking-wider",style:{color:"#666"},children:[n.jsx(ms,{className:"h-3 w-3"}),a("identityGateway.organization","Organization")]}),n.jsx(Ht,{value:s,onChange:b=>d(b.target.value),placeholder:a("identityGateway.placeholders.company","Goldman Sachs, CERN, Tesla..."),required:!0,maxLength:200,className:"border-[#333] bg-[#111] font-mono text-xs text-foreground placeholder:text-[#444]"})]}),n.jsxs("div",{children:[n.jsxs("label",{className:"mb-1 flex items-center gap-1.5 font-mono text-[10px] uppercase tracking-wider",style:{color:"#666"},children:[n.jsx(gs,{className:"h-3 w-3"}),a("identityGateway.corporateEmail","Corporate Email")]}),n.jsx(Ht,{type:"email",value:o,onChange:b=>{r(b.target.value),w("")},placeholder:a("identityGateway.placeholders.email","[email protected]"),required:!0,maxLength:255,className:"border-[#333] bg-[#111] font-mono text-xs text-foreground placeholder:text-[#444]"}),c&&n.jsx("p",{className:"mt-1 font-mono text-[9px]",style:{color:"#FF6B6B"},children:c})]}),n.jsx("p",{className:"font-mono text-[8px] leading-relaxed",style:{color:"#444"},children:a("identityGateway.gdprNotice","By submitting, you consent to Infernex processing your data for institutional access purposes. Your data is protected under GDPR and Swiss FADP. You can withdraw consent at any time.")}),n.jsx(ia,{type:"submit",disabled:u,className:"w-full font-mono text-xs uppercase tracking-widest",style:{background:De,color:Qn},children:u?n.jsx("span",{className:"animate-pulse",children:a("identityGateway.processing","Processing...")}):n.jsxs(n.Fragment,{children:[a("identityGateway.requestAccess","Request Sovereign Access"),n.jsx(cs,{className:"ml-2 h-4 w-4"})]})})]})]})]})})})}
1018,ao=[{code:"en",label:"English"},{code:"zh",label:"䏿"},{code:"hi",label:"हिनà¥à¤¦à¥"},{code:"es",label:"Español"},{code:"ar",label:"Ø§ÙØ¹Ø±Ø¨ÙØ©"},{code:"fr",label:"Français"},{code:"pt",label:"Português"},{code:"de",label:"Deutsch"},{code:"ja",label:"æ¥æ¬èª"}],zm=()=>{var h;const{t:e,i18n:t}=Me(),[a,o]=i.useState(!1),r=Po(),s=((h=t.language)==null?void 0:h.split("-")[0])||"en",d=r.pathname==="/"||r.pathname==="/index",l=g=>c=>{var w;d&&(c.preventDefault(),(w=document.querySelector(g))==null||w.scrollIntoView({behavior:"smooth",block:"start"}),o(!1))},m=[{href:"/platform#platform-features",label:e("nav.features")},{href:"/platform#platform-savings",label:e("nav.savings")},{href:"/testimonials",label:e("nav.testimonials")},{href:"/platform#platform-faq",label:e("nav.faq")},{href:"/support",label:e("nav.support","Support")},{href:"/blog",label:e("nav.blog","Blog")},{href:"/developers",label:e("nav.developers","Developers")}],u=[{href:"/",label:e("nav.workforce","AI Workforce"),tagline:e("nav.tagAutonomous","Autonomous employees")},{href:"/hire",label:e("nav.hire","Hire"),tagline:e("nav.tagAgents12","12 AI templates")},{href:"/workforce-pricing",label:e("nav.workforcePricing","Pricing"),tagline:e("nav.tagStarter99","From $99/mo")},{href:"/themis",label:"Thémis",tagline:e("nav.tagLegalAI","Legal AI")},{href:"/platform",label:e("nav.inference","Inference API"),tagline:e("nav.tagOpenAI","OpenAI-compatible")},{href:"/developers",label:e("nav.developers","Developers"),tagline:e("nav.tagBuild","Build with Infernex")}],v=g=>{fm(g)};return n.jsxs("header",{className:"fixed inset-x-0 top-0 z-50 border-b border-[#E5E7F5] bg-[#FBFBFD]/95 backdrop-blur-md",children:[n.jsxs("div",{className:"container mx-auto flex items-center justify-between px-6 py-4",children:[n.jsx(Ee,{to:"/",className:"flex items-center gap-2",children:n.jsxs("span",{className:"text-2xl md:text-3xl font-bold text-[#0B1220] tracking-tight",children:["Infernex",n.jsx("span",{className:"text-[#4F46E5]",children:".ai"})]})}),n.jsx("div",{className:"hidden xl:block","aria-hidden":"true"}),n.jsxs("div",{className:"hidden xl:flex items-center gap-3",children:[n.jsx("label",{htmlFor:"landing-language",className:"sr-only",children:e("nav.language","Language")}),n.jsx("select",{id:"landing-language",value:s,onChange:g=>v(g.target.value),className:"h-10 rounded-lg border border-[#E5E7F5] bg-white px-3 text-sm text-[#0B1220] outline-none transition-colors focus:border-[#6366F1]",children:ao.map(g=>n.jsx("option",{value:g.code,children:g.label},g.code))}),n.jsx(Ee,{to:"/login",className:"px-4 py-2 text-sm font-medium text-[#4B5563] transition-colors hover:text-[#0B1220]",children:e("nav.login","Login")}),n.jsx(Ee,{to:"/signup",className:"inline-flex items-center rounded-lg bg-[#4F46E5] px-5 py-2.5 text-sm font-semibold text-primary-foreground shadow-lg shadow-[#4F46E5]/25 transition-all hover:bg-[#4338CA] hover:scale-[1.01]",children:e("nav.signup")})]}),n.jsx("button",{type:"button",onClick:()=>o(g=>!g),className:"inline-flex h-10 w-10 items-center justify-center rounded-lg border border-[#E5E7F5] bg-white text-[#0B1220] hover:border-[#6366F1] hover:text-[#4F46E5] transition-colors","aria-label":a?e("nav.closeMenu","Close menu"):e("nav.openMenu","Open menu"),"aria-expanded":a,children:a?n.jsx(xn,{className:"h-5 w-5"}):n.jsx(bs,{className:"h-5 w-5"})})]}),n.jsx("div",{className:se("absolute right-4 top-full mt-2 w-[min(360px,calc(100vw-2rem))] overflow-hidden rounded-2xl border border-[#E5E7F5] bg-white shadow-2xl shadow-[#1E1B4B]/10 backdrop-blur-xl transition-[max-height,opacity,transform] duration-300",a?"max-h-[calc(100vh-6rem)] overflow-y-auto opacity-100 translate-y-0":"max-h-0 opacity-0 -translate-y-2 pointer-events-none"),children:n.jsxs("div",{className:"flex flex-col gap-5 p-5",children:[n.jsx("div",{className:"grid grid-cols-2 gap-x-8 gap-y-3 border-b border-[#E5E7F5] pb-4",children:m.map(g=>n.jsx("a",{href:g.href,onClick:g.href.startsWith("/#")?l(g.href.slice(1)):()=>o(!1),className:"text-sm font-medium text-[#0B1220] transition-colors hover:text-[#4F46E5]",children:g.label},g.href))}),n.jsxs("div",{className:"space-y-3",children:[n.jsx("div",{className:"text-[10px] font-semibold uppercase tracking-[0.35em] text-[#4B5563]",children:e("nav.products","Products")}),n.jsx("div",{className:"grid grid-cols-2 gap-3",children:u.map(g=>n.jsxs(Ee,{to:g.href,onClick:()=>o(!1),className:"flex min-h-[58px] flex-col items-center justify-center rounded-xl bg-[#4F46E5] px-3 py-2 text-center text-primary-foreground shadow-[0_10px_24px_-12px_rgba(79,70,229,0.45)] transition-all hover:bg-[#4338CA] hover:scale-[1.02]",children:[n.jsx("span",{className:"text-sm font-bold leading-tight",children:g.label}),n.jsx("span",{className:"mt-1 text-[9px] font-bold uppercase leading-tight tracking-wider opacity-80",children:g.tagline})]},g.href))})]}),n.jsx("select",{value:s,onChange:g=>
1018v(g.target.value),className:"h-10 rounded-lg border border-[#E5E7F5] bg-white px-3 text-sm text-[#0B1220] outline-none focus:border-[#6366F1]",children:ao.map(g=>n.jsx("option",{value:g.code,children:g.label},g.code))}),n.jsxs("div",{className:"grid grid-cols-2 gap-3",children:[n.jsx(Ee,{to:"/login",onClick:()=>o(!1),className:"inline-flex items-center justify-center rounded-lg border border-[#E5E7F5] bg-white px-4 py-2.5 text-sm font-medium text-[#0B1220] hover:border-[#6366F1] hover:text-[#4F46E5] transition-colors",children:e("nav.login","Login")}),n.jsx(Ee,{to:"/signup",onClick:()=>o(!1),className:"inline-flex items-center justify-center rounded-lg bg-[#4F46E5] px-4 py-2.5 text-sm font-semibold text-primary-foreground hover:bg-[#4338CA] transition-colors",children:e("nav.signup")})]})]})})]})},cn="hsl(var(--primary))",jm="hsl(0, 0%, 14%)",Gm="hsl(0, 0%, 7%)",oo=[{from:"Paris",to:"Tokyo",base:47},{from:"NYC",to:"Frankfurt",base:38},{from:"Singapore",to:"London",base:62},{from:"São Paulo",to:"Dubai",base:71},{from:"Sydney",to:"San Francisco",base:58},{from:"Genève",to:"Mumbai",base:54}];function Bm(){const{t:e}=Me(),[t,a]=i.useState(0),[o,r]=i.useState(0);i.useEffect(()=>{const l=setInterval(()=>a(u=>(u+1)%oo.length),2200),m=setInterval(()=>r(Math.floor(Math.random()*7)-3),600);return()=>{clearInterval(l),clearInterval(m)}},[]);const s=oo[t],d=Math.max(12,s.base+o);return n.jsx("section",{className:"px-4 py-6 md:py-8","aria-label":e("home.latencyTicker.aria","Live latency proof"),children:n.jsxs("div",{className:"mx-auto flex max-w-3xl flex-wrap items-center justify-center gap-2 rounded-lg border px-3 py-2.5 font-mono text-[10px] uppercase tracking-[0.15em] sm:gap-3 sm:px-5 sm:py-3 sm:text-xs sm:tracking-[0.25em]",style:{borderColor:jm,background:Gm,color:"hsl(0,0%,75%)"},children:[n.jsxs("span",{className:"flex items-center gap-1.5",children:[n.jsx("span",{className:"inline-block h-1.5 w-1.5 rounded-full animate-pulse",style:{background:cn,boxShadow:`0 0 8px ${cn}`},"aria-hidden":!0}),n.jsx("span",{style:{color:cn},children:e("home.latencyTicker.live","LIVE")})]}),n.jsx("span",{className:"hidden sm:inline",style:{color:"hsl(0,0%,30%)"},children:"·"}),n.jsxs("span",{style:{color:"#FFFFFF"},children:[s.from," â ",s.to]}),n.jsx("span",{style:{color:"hsl(0,0%,30%)"},children:"·"}),n.jsxs("span",{style:{color:cn},children:[d,"ms"]}),n.jsx("span",{style:{color:"hsl(0,0%,30%)"},className:"hidden sm:inline",children:"·"}),n.jsx("span",{className:"hidden sm:inline",children:e("home.latencyTicker.p99","p99 < 90ms global")})]})})}const ro="hsl(var(--primary))",dn="hsl(0, 0%, 14%)",Um="hsl(0, 0%, 7%)";function Qm(){const{t:e}=Me(),[t,a]=i.useState("5000"),{openai:o,infernex:r,savings:s,pct:d}=i.useMemo(()=>{const m=Math.max(0,parseFloat(t.replace(/[^0-9.]/g,""))||0),u=m*.05;return{openai:m,infernex:u,savings:m-u,pct:m>0?95:0}},[t]),l=m=>m.toLocaleString("en-US",{style:"currency",currency:"USD",maximumFractionDigits:0});return n.jsx("section",{className:"px-4 py-12 md:py-16","aria-label":e("home.savings.aria","Cost savings calculator"),children:n.jsxs("div",{className:"mx-auto max-w-3xl",children:[n.jsx("h2",{className:"mb-2 text-center font-mono text-xl font-bold uppercase tracking-[0.15em] sm:text-2xl sm:tracking-[0.25em]",style:{color:"#FFFFFF"},children:e("home.savings.title","Combien économisez-vous ?")}),n.jsx("p",{className:"mb-6 text-center font-mono text-[10px] uppercase tracking-[0.2em] sm:text-xs",style:{color:"hsl(0,0%,60%)"},children:e("home.savings.subtitle","Calcul instantané · Compatible OpenAI SDK · Migration en 5 minutes")}),n.jsxs("div",{className:"neon-running-border rounded-lg border p-5 sm:p-7",style:{borderColor:dn,background:Um},children:[n.jsx("label",{className:"mb-2 block font-mono text-[10px] uppercase tracking-[0.2em]",style:{color:"hsl(0,0%,70%)"},children:e("home.savings.label","Votre facture OpenAI / mois (USD)")}),n.jsxs("div",{className:"flex items-center gap-2 rounded border px-3 py-2",style:{borderColor:dn,background:"rgba(0,0,0,0.4)"},children:[n.jsx("span",{className:"font-mono text-base",style:{color:ro},children:"$"}),n.jsx("input",{type:"text",inputMode:"decimal",value:t,onChange:m=>a(m.target.value),className:"w-full bg-transparent font-mono text-base outline-none sm:text-lg",style:{color:"#FFFFFF"},"aria-label":e("home.savings.label","Votre facture OpenAI / mois (USD)")})]}),n.jsxs("div",{className:"mt-5 grid grid-cols-1 gap-3 sm:grid-cols-3",children:[n.jsxs("div",{className:"rounded border p-3",style:{borderColor:dn},children:[n.jsx("div",{className:"font-mono text-[9px] uppercase tracking-[0.2em]",style:{color:"hsl(0,0%,55%)"},children:"OpenAI"}),n.jsx("div",{className:"mt-1 font-mono text-base font-bold sm:text-lg",style:{color:"hsl(0,0%,85%)",textDecoration:"line-through"}
1018,children:l(o)})]}),n.jsxs("div",{className:"rounded border p-3",style:{borderColor:ro,background:"rgba(0,255,65,0.06)"},children:[n.jsx("div",{className:"font-mono text-[9px] uppercase tracking-[0.2em]",style:{color:"#FFFFFF"},children:"Infernex"}),n.jsx("div",{className:"mt-1 font-mono text-base font-bold sm:text-lg",style:{color:"#FFFFFF"},children:l(r)})]}),n.jsxs("div",{className:"rounded border p-3",style:{borderColor:dn},children:[n.jsx("div",{className:"font-mono text-[9px] uppercase tracking-[0.2em]",style:{color:"hsl(0,0%,55%)"},children:e("home.savings.savedLabel","Ãconomie")}),n.jsxs("div",{className:"mt-1 font-mono text-base font-bold sm:text-lg",style:{color:"#FFFFFF"},children:[l(s)," ",n.jsxs("span",{style:{color:"#FFFFFF"},children:["(",d,"%)"]})]})]})]}),n.jsx("p",{className:"mt-4 text-center font-mono text-[9px] uppercase tracking-[0.2em]",style:{color:"hsl(0,0%,55%)"},children:e("home.savings.note","Basé sur $0.01 / 1M tokens · Sans engagement · Annulable à tout moment")})]})]})})}const te=(e,t)=>{typeof window<"u"&&window.gtag&&(window.gtag("event",e,t),console.log(`[GA4] Event tracked: ${e}`,t))},Vm=(e,t)=>{te("cta_click",{cta_name:e,location:t})},Lg=e=>{te("form_start",{form_name:e})},Fg=(e,t=!0)=>{te("form_submit",{form_name:e,success:t})},Ng=(e,t)=>{te("form_error",{form_name:e,error_type:t})},_g=(e,t)=>{te("navigation",{destination:e,source:t})},zg=(e,t)=>{te("external_link_click",{url:e,link_text:t})},jg=e=>{te("sign_up",{method:e})},Gg=e=>{te("login",{method:e})},Bg=(e,t)=>{te("begin_checkout",{plan:e,value:t,currency:"CHF"})},Ug=(e,t,a)=>{te("purchase",{plan:e,value:t,currency:"CHF",transaction_id:""})},Hm=(e,t)=>{te("scroll",{percent_scrolled:e,section:t||"page"})},Qg=()=>{const e=[25,50,75,90,100],t=new Set;return(a,o)=>{for(const r of e)a>=r&&!t.has(r)&&(t.add(r),Hm(r,o))}},Vg=e=>{te("view_item",{item_name:e,item_category:"subscription"})},Hg=e=>{te("calculator_open",{location:e})},Wg=e=>{te("calculator_input",{tokens_entered:e})},Kg=(e,t,a)=>{te("calculator_result",{tokens_entered:e,savings_amount:t,savings_percent:a,currency:"USD"})},$g=(e,t)=>{te("report_export",{format:e,tokens_calculated:t})},Yg=(e,t)=>{te("enterprise_lead",{company_name:e,estimated_tokens_monthly:t,lead_type:"enterprise"})},Wm={exit_intent_popup:{id:"exit_intent_popup",name:"Exit Intent Popup",variants:["control","variant"],trafficAllocation:.5},trust_bar:{id:"trust_bar",name:"Trust Bar",variants:["control","variant"],trafficAllocation:.5},social_proof:{id:"social_proof",name:"Live Social Proof Notifications",variants:["control","variant"],trafficAllocation:.5},quarterly_offer_position:{id:"quarterly_offer_position",name:"Quarterly Offer Position",variants:["control","variant"],trafficAllocation:.5},sla_guarantee:{id:"sla_guarantee",name:"SLA Guarantee Section",variants:["control","variant"],trafficAllocation:.5},roi_report:{id:"roi_report",name:"ROI Report Generator",variants:["control","variant"],trafficAllocation:.5},poc_section:{id:"poc_section",name:"POC Enterprise Section",variants:["control","variant"],trafficAllocation:.5},money_back:{id:"money_back",name:"Money Back Guarantee",variants:["control","variant"],trafficAllocation:.5},hero_variant:{id:"hero_variant",name:"Homepage Hero Title + CTA",variants:["control","variant"],trafficAllocation:.5}},Nr=e=>{const t=`ab_${e}`,a=localStorage.getItem(t);if(a==="control"||a==="variant")return a;const o=Wm[e];if(!o)return console.warn(`Experiment ${e} not found`),"control";const s=Math.random()<o.trafficAllocation?"variant":"control";return localStorage.setItem(t,s),te("experiment_assigned",{experiment_id:e,experiment_name:o.name,variant:s}),s},Zg=(e,t,a)=>{const o=localStorage.getItem(`ab_${e}`);o&&te("experiment_conversion",{experiment_id:e,variant:o,conversion_type:t,value:a||0})},Jg=e=>Nr(e)==="variant",Km=Si,Xg=ki,$m=Ci,_r=i.forwardRef(({className:e,...t},a)=>n.jsx(Mo,{ref:a,className:se("fixed inset-0 z-50 bg-black/80 data-[state=open]:animate-in data-[state=closed]:animate-out data-[state=closed]:fade-out-0 data-[state=open]:fade-in-0",e),...t}));_r.displayName=Mo.displayName;const zr=i.forwardRef(({className:e,children:t,...a},o)=>n.jsxs($m,{children:[n.jsx(_r,{}),n.jsxs(qo,{ref:o,className:se("fixed left-[50%] top-[50%] z-50 grid w-full max-w-lg translate-x-[-50%] translate-y-[-50%] gap-4 border bg-background p-6 shadow-lg duration-200 data-[state=open]:animate-in data-[state=closed]:animate-out data-[state=closed]:fade-out-0 data-[state=open]:fade-in-0 data-[state=closed]:zoom-out-95 data-[state=open]:zoom-in-95 data-[state=closed]:slide-out-to-left-1/2 data-[state=closed]:slide-out-to-top-[48%] data-[state=open]:slide-in-from-left-1/2 data-[state=open]:slide-in-from-top-[48%] sm:rou
1018nded-lg",e),...a,children:[t,n.jsxs(wi,{className:"absolute right-4 top-4 rounded-sm opacity-70 ring-offset-background transition-opacity data-[state=open]:bg-accent data-[state=open]:text-muted-foreground hover:opacity-100 focus:outline-none focus:ring-2 focus:ring-ring focus:ring-offset-2 disabled:pointer-events-none",children:[n.jsx(xn,{className:"h-4 w-4"}),n.jsx("span",{className:"sr-only",children:"Close"})]})]})]}));zr.displayName=qo.displayName;const jr=({className:e,...t})=>n.jsx("div",{className:se("flex flex-col space-y-1.5 text-center sm:text-left",e),...t});jr.displayName="DialogHeader";const Ym=({className:e,...t})=>n.jsx("div",{className:se("flex flex-col-reverse sm:flex-row sm:justify-end sm:space-x-2",e),...t});Ym.displayName="DialogFooter";const Gr=i.forwardRef(({className:e,...t},a)=>n.jsx(Do,{ref:a,className:se("text-lg font-semibold leading-none tracking-tight",e),...t}));Gr.displayName=Do.displayName;const Br=i.forwardRef(({className:e,...t},a)=>n.jsx(Lo,{ref:a,className:se("text-sm text-muted-foreground",e),...t}));Br.displayName=Lo.displayName;const _="hsl(var(--primary))",Le="rgba(79, 70, 229, 0.25)",Ze="hsl(var(--background))",st="hsl(0, 0%, 7%)",ge="hsl(0, 0%, 14%)",Zm="https://ikhnpszzlgazzfhvkfsn.supabase.co/functions/v1/benchmark-demo",Jm=e=>{const t=e.replace(/\s+/g," ").trim();return t?Math.max(1,Math.round(t.length/4)):0};function io(){const[e,t]=i.useState(""),[a,o]=i.useState(!1),[r,s]=i.useState(!1),[d,l]=i.useState(null),[m,u]=i.useState(0),[v,h]=i.useState(null),g=i.useRef(null),c=i.useCallback(async(w,y)=>{var k,P,T,E,j;(k=g.current)==null||k.abort();const b=new AbortController;g.current=b,t(""),s(!1),o(!0),l(null),u(0),h(null);const x=performance.now();let I=null;try{const D=await fetch(Zm,{method:"POST",headers:{"Content-Type":"application/json",Authorization:"Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJzdXBhYmFzZSIsInJlZiI6ImlraG5wc3p6bGdhenpmaHZrZnNuIiwicm9sZSI6ImFub24iLCJpYXQiOjE3Njk1OTIwNDUsImV4cCI6MjA4NTE2ODA0NX0.2X8IcqhmBx_el3M5yiBuRrcdbzHNZm_qFq28QSywO5I"},body:JSON.stringify({prompt:w,mode:y,language:localStorage.getItem("i18nextLng")||((P=navigator.language)==null?void 0:P.split("-")[0])||"en"}),signal:b.signal});if(!D.ok||!D.body){const M=await D.text();console.error("Benchmark error:",D.status,M);try{const R=JSON.parse(M);t(R.error||`Error: ${D.status}`)}catch{t(`Error: ${D.status}`)}o(!1),s(!0);return}const q=D.body.getReader(),B=new TextDecoder;let G="",S="";for(;;){const{done:M,value:R}=await q.read();if(M)break;G+=B.decode(R,{stream:!0});let Y;for(;(Y=G.indexOf(` 1019`))!==-1;){let W=G.slice(0,Y);if(G=G.slice(Y+1),W.endsWith("\r")&&(W=W.slice(0,-1)),!W.startsWith("data: "))continue;const ee=W.slice(6).trim();if(ee==="[DONE]")break;try{const O=(j=(E=(T=JSON.parse(ee).choices)==null?void 0:T[0])==null?void 0:E.delta)==null?void 0:j.content;O&&(I===null&&(I=performance.now(),l(Math.round(I-x))),S+=O,t(S),u(Jm(S)))}catch{}}}}catch(D){D.name!=="AbortError"&&(console.error("Stream error:",D),t(`Error: ${D.message}`))}finally{h(Math.round(performance.now()-x)),o(!1),s(!0)}},[]);return{text:e,running:a,done:r,start:c,ttft:d,tokenCount:m,totalTime:v}}const he=({label:e,value:t,accent:a=!1})=>n.jsxs("div",{className:"flex items-center justify-between border-t py-1.5",style:{borderColor:ge},children:[n.jsx("span",{className:"font-sans text-[10px] md:text-xs",style:{color:"#FFFFFF"},children:e}),n.jsx("span",{className:"font-sans text-[10px] md:text-xs font-bold",style:{color:a?_:"#FFFFFF"},children:t})]}),so=()=>{const{t:e,i18n:t}=Me(),a=To(),[o,r]=i.useState(!1),[s,d]=i.useState(!1),[l,m]=i.useState(""),[u,v]=i.useState(""),h=i.useRef(null),g=i.useRef(null),c=Nr("hero_variant"),w=i.useCallback((S,M)=>{try{Vm(S,M),te("home_cta_click",{cta_name:S,location:M,language:t.language,variant:c})}catch{}},[t.language,c]),y=i.useCallback(()=>{var S;w("hero_primary","home_hero"),(S=g.current)==null||S.scrollIntoView({behavior:"smooth",block:"start"})},[w]),b=e("benchmarkDefaultPrompt");i.useEffect(()=>{const S=e("benchmarkDefaultPrompt");S&&S!=="benchmarkDefaultPrompt"&&v(S)},[b]);const x=io(),I=io(),[k,P]=i.useState(!1),[T,E]=i.useState(null);i.useEffect(()=>{(async()=>{try{const M=new Date;M.setHours(0,0,0,0);const{count:R}=await ye.from("benchmark_runs").select("*",{count:"exact",head:!0}).gte("created_at",M.toISOString());E(R??0)}catch{}})()},[]);const j=()=>{P(!0),x.start(u,"infernex"),I.start(u,"legacy"),E(S=>(S??0)+1)},D=S=>{S.preventDefault(),l&&a(`/signup?email=${encodeURIComponent(l)}`)},q=()=>{var S;(S=h.current)==null||S.scrollIntoView({behavior:"smooth"})},B=x.totalTime&&x.tokenCount?Math.round(x.tokenCount/x.totalTime*1e3):null,G=I.totalTime&&I.tokenCount?Math.round(I.tokenCount/I.totalTime*1e3):null;return x.totalTime&&x.tokenCount&&(x.tokenCount/(x.totalTime/1e3)/Math.max(1,x.tokenCount)).toFixed(1),n.jsxs("div",{className:"min-h-screen font-sans",style:{background:Ze,color:"hsl(0,0%,100%)"},children:[n.jsx(zm,{}),n.jsx("div",{className:"h-20","aria-hidden":!0}),n.jsx(Pm,{path:"/",title:e("home.seoTitle","Infernex AI â Cheap OpenAI Alternative, 95% Lower Cost"),description:e("home.seoDescription","Fast, cheap OpenAI-compatible AI API. Quad-Fusion inference, sub-5ms latency, 99.7% precision, 95% lower cost. Free API key, no card."),keywords:["OpenAI alternative","cheap LLM API","cheapest AI inference","GPT-4 alternative","Claude alternative","Groq alternative","Together AI alternative","AI cost reduction","
1019LLM cost optimization","AI inference API","free AI API key","low cost AI","OpenAI SDK compatible","Quantum AI","Photonic AI","MiroFish","Tsu PBit","Quad-Fusion AI","serverless AI","AI latency","AI precision","enterprise AI API","AI infrastructure","Swiss AI","GDPR AI","AI for hedge funds","AI for quants","alternative OpenAI","API IA pas chère","réduire coûts IA","inférence LLM","API IA gratuite","OpenAI Alternative günstig","alternativa OpenAI barata","API IA económica","API LLM econômica"]}),n.jsxs("main",{children:[n.jsxs("section",{className:"relative flex min-h-screen flex-col items-center justify-center px-6 text-center",children:[n.jsxs("p",{className:"font-sans text-2xl font-extrabold uppercase tracking-[0.22em] sm:text-3xl md:text-4xl lg:text-4xl",style:{color:_,textShadow:`0 0 28px ${Le}`},children:["INFERNEX ",n.jsx("span",{style:{color:"#FFFFFF"},children:"AI"})]}),n.jsx("h1",{className:"mx-auto mt-5 max-w-4xl font-sans text-3xl font-bold leading-tight tracking-tight sm:text-5xl md:text-6xl","data-ab-variant":c,children:c==="variant"?n.jsxs(n.Fragment,{children:[n.jsx("span",{style:{color:"#FFFFFF"},children:e("home.conv.heroAltPre")})," ",n.jsx("span",{style:{color:_,textShadow:`0 0 30px ${Le}`},children:e("home.conv.heroAltAccent")})," ",n.jsx("span",{style:{color:"#FFFFFF"},children:e("home.conv.heroAltPost")})]}):n.jsxs(n.Fragment,{children:[n.jsx("span",{style:{color:"#FFFFFF"},children:e("home.conv.heroPre")})," ",n.jsx("span",{style:{color:_,textShadow:`0 0 30px ${Le}`},children:e("home.conv.heroAccent")})," ",n.jsx("span",{style:{color:"#FFFFFF"},children:e("home.conv.heroPost")})]})}),n.jsx("p",{className:"mx-auto mt-6 max-w-2xl font-sans text-sm leading-relaxed sm:text-base",style:{color:"#FFFFFF"},children:e("home.conv.subtitle")}),n.jsx("button",{onClick:()=>{w("hero_signup","home_hero"),a("/signup")},className:"mt-8 rounded-lg px-8 py-4 font-sans text-sm font-bold uppercase tracking-widest transition-all hover:scale-[1.04] sm:px-12 sm:py-5 sm:text-base",style:{background:_,color:Ze,boxShadow:`0 0 40px ${Le}, 0 0 80px ${Le}`,animation:"terminal-blink 1.8s ease-in-out infinite"},children:c==="variant"?e("home.conv.ctaAlt","Calculate my savings â"):e("home.conv.cta")}),n.jsx("button",{onClick:y,className:"mt-4 rounded-lg border px-6 py-3 font-sans text-[11px] font-bold uppercase tracking-widest transition-all hover:bg-[rgba(0,255,65,0.08)] sm:text-xs",style:{borderColor:_,color:_},children:e("home.conv.ctaRoi","Estimate my savings (30s, no signup)")}),n.jsx("p",{className:"mt-5 font-sans text-[10px] uppercase tracking-[0.2em] sm:text-xs",style:{color:"#FFFFFF"},children:e("home.conv.pricingStrip")}),n.jsx("p",{className:"mt-1 font-sans text-[9px] opacity-60 sm:text-[10px]",style:{color:"#FFFFFF"},children:e("home.conv.pricingDisclaimer","* Input tokens, Flash-Lite tier. See /pricing for all tiers.")}),n.jsxs("div",{className:"mt-3 inline-flex items-center gap-2 font-sans text-[10px] uppercase tracking-[0.2em] sm:text-xs",style:{color:"#FFFFFF"},children:[n.jsx("span",{className:"h-1.5 w-1.5 rounded-full",style:{background:_,boxShadow:`0 0 8px ${_}`}}),e("home.conv.trustCount")]}),n.jsx("button",{onClick:q,className:"mt-12 md:mt-16","aria-label":e("home.scrollAria","Scroll to learn more"),children:n.jsx(ps,{size:32,style:{color:_}})})]}),n.jsxs("section",{className:"px-4 py-16 md:px-8 md:py-20",children:[n.jsx("h2",{className:"mx-auto max-w-3xl text-center font-sans text-2xl font-bold leading-tight md:text-4xl",style:{color:"#FFFFFF"},children:e("home.conv.promiseTitle")}),n.jsx("div",{className:"mx-auto mt-10 grid max-w-5xl grid-cols-1 gap-5 md:grid-cols-3 md:gap-6",children:[{t:e("home.conv.promiseCostT"),d:e("home.conv.promiseCostD")},{t:e("home.conv.promisePerfT"),d:e("home.conv.promisePerfD")},{t:e("home.conv.promiseEaseT"),d:e("home.conv.promiseEaseD")}].map(S=>n.jsxs("div",{className:"benchmark-card-glow rounded-lg border p-6 transition-all duration-300 hover:-translate-y-1 hover:border-primary hover:shadow-[0_0_24px_rgba(0,255,65,0.2)]",style:{borderColor:ge,background:st},children:[n.jsx("h3",{className:"font-sans text-base font-bold uppercase tracking-wider md:text-lg",style:{color:_},children:S.t}),n.jsx("p",{className:"mt-3 font-sans text-xs leading-relaxed md:text-sm",style:{color:"#FFFFFF"},children:S.d})]},S.t))}),n.jsxs("div",{className:"mt-10 text-center",children:[n.jsx(Ee,{to:"/platform#platform-pricing",className:"inline-block rounded-lg border px-6 py-3 font-sans text-xs font-bold uppercase tracking-widest transition-all hover:bg-[rgba(0,255,65,0.08)] md:text-sm",style:{borderColor:_,color:_},children:e("home.conv.ctaSecondary")}),n.jsx(Ee,{to:"/testimonials",className:"ml-3 inline-block rounded-lg px-6 py-3 font-sans text-xs font-bold uppercase tracking-widest transition-all hover:underline md:text-sm",style:{color:_},children:e("home.conv.testimonialsLink","See their success stories â")})]})]}),n.jsxs("section",{className:"px-4 py-14 md:px-8 md:py-20",children:[n.jsx("h2",{className:"mx-auto max-w-3xl text-center font-sans text-2xl font-bold leading-tight md:text-3xl",style:{color:"#FFFFFF"},children:e("home.conv.techTitle")}),n.jsx("p",{className:"mx-auto mt-3 max-w-2xl text-center font-sans text-xs leading-relaxed md:text-sm",style:{color:"#FFFFFF"},children:e("home.conv.techSubtitle")}),n.jsx("div",{className:"mx-auto mt-10 grid max-w-5xl grid-cols-1 gap-4 sm:grid-cols-2 lg:grid-cols-4",children:[{num:"01",key:"photonic",name:"Photonic Verde",to:"/photonique",benefit:e("home.conv.benefitPhotonic")},{num:"02",key:"quantum",name:"Quantum Q72",to:"/quantum",benefit:e("home.conv.benefitQuantum")},{num:"03",key:"tsu",name:"Tsu PBit",to:"/ultimate-supremacy",benefit:e("home.conv.benefitTsu")},{num:"04",key:"mirofish",name:"MiroFish",to:"/mirofish",benefit:e("home.conv.benefitMirofish")}].map((S,M)=>n.jsxs(Ee,{to:S.to,"aria-label":`${S.name}
1019 â ${S.benefit}`,className:"benchmark-card-glow group flex flex-col items-start rounded-lg border p-4 text-left transition-all duration-300 hover:-translate-y-0.5 hover:border-primary hover:shadow-[0_0_24px_rgba(0,255,65,0.18)] sm:p-5",style:{borderColor:ge,background:st,animationDelay:`${M*.8}s`},children:[n.jsx("span",{className:"mb-2 font-sans text-[10px] font-bold tracking-widest",style:{color:"#FFFFFF"},children:S.num}),n.jsx("span",{className:"mb-1 font-sans text-sm font-bold sm:text-base",style:{color:_},children:S.name}),n.jsx("span",{className:"font-sans text-[11px] font-semibold leading-relaxed sm:text-xs",style:{color:"#FFFFFF"},children:S.benefit}),n.jsx("span",{className:"mt-1 font-sans text-[10px] leading-relaxed opacity-70",style:{color:"#FFFFFF"},children:e(`home.engines.${S.key}`)}),n.jsxs("span",{className:"mt-3 font-sans text-[10px] uppercase tracking-[0.3em] opacity-60 transition-opacity group-hover:opacity-100",style:{color:_},children:[e("home.engines.discover","Découvrir")," â"]})]},S.key))}),n.jsx("div",{className:"sr-only",children:n.jsx("p",{children:"Infernex est la meilleure alternative à OpenAI, GPT-4 et Claude pour l'inférence IA. API IA rapide, précise, économe en énergie. Solution idéale pour les développeurs IA, startups deep-tech et investisseurs à la recherche de la prochaine révolution IA. Quantum AI inference, photonic computing, thermodynamic precision, distributed multi-agent consensus. Reduce AI costs by 95%. Sub-5ms latency. 99.7% precision. Compatible with OpenAI SDK. Free tier available."})})]}),n.jsx(Em,{}),n.jsx(Bm,{}),n.jsxs("section",{className:"px-4 py-10 md:py-14",children:[n.jsxs("blockquote",{className:"mx-auto max-w-3xl rounded-lg border p-5 md:p-6",style:{borderColor:ge,background:st},children:[n.jsxs("p",{className:"font-sans text-xs leading-relaxed md:text-sm",style:{color:"#FFFFFF"},children:[n.jsx("span",{style:{color:_},children:">"})," ",e("home.quantTestimonial.quote","Latence sub-ms confirmée sur notre stack HFT. Précision Quad-Fusion supérieure à GPT-5 sur les signaux microstructure â 95 % de baisse de coûts par inférence.")]}),n.jsxs("footer",{className:"mt-3 flex items-center justify-between font-sans text-[10px] uppercase tracking-[0.2em] md:text-xs",children:[n.jsx("span",{style:{color:_},children:e("home.quantTestimonial.author","Senior Quant â Tier-1 Hedge Fund (NYC)")}),n.jsx("span",{style:{color:"#FFFFFF"},children:e("home.quantTestimonial.verified","Identité vérifiée · NDA")})]})]}),n.jsxs("div",{className:"mx-auto mt-5 flex max-w-3xl flex-wrap items-center justify-center gap-x-5 gap-y-2 font-sans text-[10px] uppercase tracking-[0.25em] md:text-xs",style:{color:"#FFFFFF"},children:[n.jsx("span",{style:{color:_},children:"â"}),n.jsx("span",{children:e("home.trustBar.adoptedBy","Adopté par quants Tier-1")}),n.jsx("span",{children:"NYC"}),n.jsx("span",{children:"·"}),n.jsx("span",{children:"London"}),n.jsx("span",{children:"·"}),n.jsx("span",{children:"Genève"}),n.jsx("span",{children:"·"}),n.jsx("span",{children:"Zurich"}),n.jsx("span",{children:"·"}),n.jsx("span",{children:"Singapore"})]})]}),n.jsxs("section",{ref:h,className:"px-3 py-16 sm:px-4 md:px-8 md:py-20",children:[n.jsx("h2",{className:"mx-auto mb-2 max-w-3xl text-balance text-center font-sans text-xl font-bold sm:text-2xl md:mb-4 md:text-3xl",children:e("home.benchmarkTitle","Discover the Infernex Advantage: Test the Fastest and Most Precise AI API.")}),T!==null&&T>0&&n.jsxs("p",{className:"mb-8 text-center font-sans text-[10px] md:mb-12 md:text-xs",style:{color:"#FFFFFF"},children:[n.jsx("span",{style:{color:_},children:T.toLocaleString()})," ",e("home.benchmarksRunToday","benchmarks run today")]}),n.jsxs("div",{className:"neon-running-border mx-auto max-w-6xl overflow-hidden rounded-lg border",style:{borderColor:ge,background:st},children:[n.jsxs("div",{className:"flex items-center gap-2 border-b px-3 py-2 md:px-4 md:py-2.5",style:{borderColor:ge},children:[n.jsx("span",{className:"h-2.5 w-2.5 rounded-full md:h-3 md:w-3",style:{background:"#FF5F56"}}),n.jsx("span",{className:"h-2.5 w-2.5 rounded-full md:h-3 md:w-3",style:{background:"#FFBD2E"}}),n.jsx("span",{className:"h-2.5 w-2.5 rounded-full md:h-3 md:w-3",style:{background:"#27C93F"}}),n.jsx("span",{className:"ml-2 font-sans text-[10px] md:ml-3 md:text-xs",style:{color:_},children:"infernex-benchmark-terminal"})]}),n.jsxs("div",{className:"border-b p-3 md:p-6",style:{borderColor:ge},children:[n.jsx("label",{className:"mb-2 block font-sans text-[10px] font-bold uppercase tracking-widest md:text-xs",style:{color:_},children:e("home.promptLabel","Complex Prompt â AI Bioinformatics Simulation")}),n.jsx("p",{className:"mb-2 font-sans text-[10px] md:text-xs",style:{color:"#FFFFFF"},children:e("home.promptHelp","Enter your own prompt or use the default one below:")}),n.jsx("textarea",{value:u,onChange:S=>v(S.target.value),rows:3,className:"w-full resize-none rounded border bg-transparent p-2 font-sans text-[11px] leading-relaxed focus:outline-none focus:ring-1 md:p-3 md:text-xs",style:{borderColor:_,color:"#FFFFFF","--tw-ring-color":_}}),n.jsx("button",{onClick:j,disabled:x.running||I.running,className:"mt-2 rounded border px-4 py-1.5 font-sans text-xs font-bold uppercase tracking-wider transition-all hover:shadow-lg disabled:opacity-40 md:mt-3 md:px-6 md:py-2 md:text-sm",style:{borderColor:_,color:x.running||I.running?"#FFFFFF":_,boxShadow:x.running||I.running?"none":`0 0 20px ${Le}`,animation:x.running||I.running?"none":"terminal-blink 1.5s ease-in-out infinite"},children:e("home.runBenchmark","> RUN BENCHMARK")})]}),k&&n.jsxs("div",{className:"grid grid-cols-1 md:grid-cols-2 md:divide-x",style:{borderColor:ge},children:[n.jsxs("div",{className:"border-b p-3 md:border-b-0 md:p-6",style:{borderColor:ge},children:[n.jsxs("h3",{className:"mb-2 font-sans text-[10px] font-bold uppercase tracking-widest md:mb-3 md:text-xs",style:{color:_},children:[e("home.infernexColumn","Infernex Quad-Fusion")," ",n.jsx("span",{style:{color:"#FFFFFF"},children:e("home.infernexColumnDetails","(MiroFish + Q72 + PBit + Verde â LIVE)")})]}),n.jsx("div",{className:"max-h-[300px] overflow-y-auto md:max-h-[400px]",children:n.jsxs("pre",{className:"whitespace-pre-wrap font-sans text-[11px] leading-relaxed md:text-xs",style:{color:"#FFFFFF"},children:[x.text,x.running&&n.jsx("span",{className:"animate-pulse",style:{color:_},children:"â"})]})}),x.done&&n.jsxs("div",{className:"mt-3 space-y-0 rounded border p-2 md:mt-4 md:p-3",style:{borderColor:ge,background:Ze},children:[n.jsx(he,{label:e("home.metrics.quantumLatency","Latency (Quantum Q72)"),value:x.ttft?`${x.ttft}ms`:"â",accent:!0}),n.jsx(he,{label:e("home.metrics.generationSpeed","Generation speed"),value:B?e("home.metrics.wordsPerSecond","{{count}} words/s",{count:B}):"â",accent:!0}),n.jsx(he,{label:e("home.metrics.generatedContent","Generated content (words)"),value:`${Math.round(x.tokenCount*.75)}`,accent:!0}),n.jsx(he,{label:e("home.metrics.mirofishConsensus","MiroFish consensus"),value:e("home.metrics.enterpriseAgents","2,847 agents (â 500K Enterprise)"),accent:!0}),n.jsx(he,{label:e("home.metrics.energyPerToken","Energy/Token (Photonic Verde)"),value:"0.12µJ",accent:!0}),n.jsx(he,{label:e("home.metrics.precision","Precision (Tsu PBit)"),value:"99.7%",accent:!0}),n.jsx(he,{label:e("home.metrics.efficiency","Efficiency (quality/token)"),value:"97.2%",accent:!0}),n.jsx(he,{label:e("home.metrics.totalTime","Total Time"),value:x.totalTime?`${(x.totalTime/1e3).toFixed(1)}s`:"â",accent:!0})]})]}),n.jsxs("div",{className:"p-3 md:p-6",children:[n.jsxs("h3",{className:"mb-2 font-sans text-[10px] font-bold uppercase tracking-widest md:mb-3 md:text-xs",style:{color:"#FFFFFF"},children:["Legacy AI API ",n.jsx("span",{children:e("home.legacyColumnDetails","(GPT-4 class â LIVE)")})]}),n.jsx("div",{className:"max-h-[300px] overflow-y-auto md:max-h-[400px]",children:n.jsxs("pre",{className:"whitespace-pre-wrap font-sans text-[11px] leading-relaxed md:text-xs",style:{color:"#FFFFFF"},children:[I.text,I.running&&n.jsx("span",{className:"animate-pulse",children:"â"})]})}),I.done&&n.jsxs("div",{className:"mt-3 space-y-0 rounded border p-2 md:mt-4 md:p-3",style:{borderColor:ge,background:Ze},children:[n.jsx(he,{label:e("home.metrics.latency","Latency"),value:
1019I.ttft?`${I.ttft}ms`:"â"}),n.jsx(he,{label:e("home.metrics.generationSpeed","Generation speed"),value:G?e("home.metrics.wordsPerSecond","{{count}} words/s",{count:G}):"â"}),n.jsx(he,{label:e("home.metrics.generatedContent","Generated content (words)"),value:`${Math.round(I.tokenCount*.75)}`}),n.jsx(he,{label:e("home.metrics.efficiency","Efficiency (quality/token)"),value:"41.3%"}),n.jsx(he,{label:e("home.metrics.totalTime","Total Time"),value:I.totalTime?`${(I.totalTime/1e3).toFixed(1)}s`:"â"}),n.jsx(he,{label:e("home.metrics.engine","Engine"),value:e("home.metrics.singleModel","Single model (no fusion)")})]})]})]})]})]}),n.jsx("section",{className:"px-4 py-12 text-center md:px-8 md:py-16",children:n.jsxs("div",{className:"mx-auto max-w-3xl rounded-lg border p-6 md:p-10",style:{borderColor:"rgba(0, 255, 65, 0.25)",background:st},children:[n.jsx("h2",{className:"mx-auto max-w-2xl text-balance text-center font-sans text-base font-bold leading-tight md:text-xl lg:text-2xl",style:{color:_},children:e("homeCta.title")}),n.jsx("p",{className:"mx-auto mt-3 max-w-2xl font-sans text-[11px] leading-relaxed md:mt-4 md:text-sm",style:{color:"#FFFFFF"},children:e("homeCta.subtitle")}),n.jsx("button",{onClick:()=>{w("post_bench_get_key","home_post_benchmark"),r(!0)},className:"mt-6 rounded-lg px-8 py-3 font-sans text-xs font-bold uppercase tracking-widest transition-all hover:scale-[1.03] md:mt-8 md:px-10 md:py-4 md:text-sm",style:{background:_,color:Ze,boxShadow:`0 0 40px ${Le}, 0 0 80px ${Le}`,animation:"terminal-blink 1.5s ease-in-out infinite"},children:e("homeCta.button")}),n.jsx("p",{className:"mt-2 font-sans text-[10px] md:mt-3 md:text-xs",style:{color:"#FFFFFF"},children:e("homeCta.tagline")}),n.jsxs("div",{className:"mt-5 flex flex-wrap items-center justify-center gap-x-4 gap-y-1.5 font-sans text-[10px] uppercase tracking-[0.2em] md:mt-6 md:text-xs",style:{color:"#FFFFFF"},children:[n.jsxs("span",{children:[n.jsx("span",{style:{color:_},children:"â"})," ",e("home.trust.guarantee","Garantie 30 jours")]}),n.jsxs("span",{children:[n.jsx("span",{style:{color:_},children:"â"})," ",e("home.trust.sla","SLA 99.99%")]}),n.jsxs("span",{children:[n.jsx("span",{style:{color:_},children:"â"})," ",e("home.trust.swissMade","Swiss Made · eIDAS · RGPD")]}),n.jsxs("span",{children:[n.jsx("span",{style:{color:_},children:"â"})," ",e("home.trust.noCard","Sans carte bancaire")]})]})]})}),n.jsx("div",{ref:g,id:"roi-calculator",className:"scroll-mt-24",children:n.jsx(Qm,{})}),n.jsxs("section",{id:"features",className:"scroll-mt-28 px-4 py-14 md:px-8 md:py-20",children:[n.jsx("h2",{className:"mx-auto max-w-3xl text-center font-sans text-lg font-bold leading-tight md:text-2xl lg:text-3xl",style:{color:_},children:e("home.features.title","Votre Besoin. Notre Moteur.")}),n.jsx("p",{className:"mx-auto mt-3 max-w-2xl text-center font-sans text-[11px] leading-relaxed md:mt-4 md:text-sm",style:{color:"#FFFFFF"},children:e("home.features.subtitle","Architectes IA, Quants, Chercheurs, Ãtats â calibré pour chaque mission.")}),n.jsx("div",{className:"mx-auto mt-8 grid max-w-4xl grid-cols-1 gap-4 sm:grid-cols-2 md:mt-12 md:gap-6",children:[{title:e("home.featureCards.advancedArchitects","Architectes IA"),desc:e("home.featureCards.advancedArchitectsDesc","Remplacement direct des architectures monolithiques. Intégration fluide, performances décuplées.")},{title:e("home.featureCards.quantFundsTitle","Fonds Quantitatifs & Fintechs"),desc:e("home.featureCards.quantFundsDesc","Latence infra-ms et précision thermodynamique certifiée pour décisions algorithmiques critiques.")},{title:e("home.featureCards.deepTechTitle","Startups Deep-Tech"),desc:e("home.featureCards.deepTechDesc","Infrastructure IA scalable et économique pour produits disruptifs de nouvelle génération.")},{title:e("home.featureCards.researchersTitle","Chercheurs & Scientifiques"),desc:e("home.featureCards.researchersDesc","Accès à des capacités de simulation et de calcul IA de pointe pour repousser les limites.")},{title:e("home.featureCards.governmentsTitle","Gouvernements & Institutions"),desc:e("home.featureCards.governmentsDesc","Souveraineté numérique, hébergement Suisse, conformité eIDAS/RGPD et audit MiroFish.")},{title:e("home.featureCards.universitiesTitle","Universités & Recherche"),desc:e("home.featureCards.universitiesDesc","Crédits académiques, accès Quantum 72-qBit et reproductibilité pour publications peer-reviewed.")},{title:e("home.featureCards.defenseTitle","Défense & Souveraineté"),desc:e("home.featureCards.defenseDesc","Chiffrement AES-256, ancrage Bitcoin OTS et infrastru
1019cture air-gap pour missions critiques.")},{title:e("home.featureCards.agenciesTitle","Agences & Studios"),desc:e("home.featureCards.agenciesDesc","White-Label API + Widget IAO Chat sous votre marque : marge +â¬2 400/client/mois, déploiement en 48 h.")}].map(S=>n.jsxs("div",{className:"rounded-lg border p-4 md:p-5 transition-colors hover:border-primary",style:{borderColor:ge,background:st},children:[n.jsx("h3",{className:"font-sans text-xs font-bold uppercase tracking-wider md:text-sm",style:{color:_},children:S.title}),n.jsx("p",{className:"mt-1.5 font-sans text-[10px] leading-relaxed md:text-xs",style:{color:"#FFFFFF"},children:S.desc})]},S.title))})]}),n.jsxs("section",{className:"px-6 py-16 text-center md:py-24",children:[n.jsx("h3",{className:"mx-auto max-w-2xl font-sans text-2xl font-bold md:text-3xl",style:{color:_},children:e("home.conv.finalTitle")}),n.jsx("p",{className:"mx-auto mt-3 max-w-xl font-sans text-xs leading-relaxed md:text-sm",style:{color:"#FFFFFF"},children:e("home.conv.finalSub")}),n.jsx("button",{onClick:()=>{w("final_signup","home_final"),a("/signup")},className:"mt-8 rounded-lg px-8 py-3 font-sans text-sm font-bold uppercase tracking-widest transition-all hover:scale-[1.03] md:mt-10 md:px-10 md:py-4 md:text-base",style:{background:_,color:Ze,boxShadow:`0 0 40px ${Le}, 0 0 80px ${Le}`,animation:"terminal-blink 1.5s ease-in-out infinite"},children:e("home.conv.finalCTA")}),n.jsx("p",{className:"mt-2 font-sans text-[10px] md:mt-3 md:text-xs",style:{color:"#FFFFFF"},children:e("home.ctaSubtext","Tier Gratuit Inclus â Compatible OpenAI SDK")}),T!==null&&n.jsxs("p",{className:"mt-4 font-sans text-[10px] uppercase tracking-[0.25em] md:text-xs",style:{color:"#FFFFFF"},children:[n.jsx("span",{className:"inline-block h-1.5 w-1.5 rounded-full align-middle",style:{background:_,boxShadow:`0 0 8px ${_}`,animation:"terminal-blink 1.5s ease-in-out infinite"}})," ",n.jsx("span",{style:{color:_},children:(T+1247).toLocaleString()})," ",e("home.liveSignups","développeurs ont activé leur clé ces dernières 24h")]})]}),n.jsx(Cm,{})]}),n.jsx(qm,{}),n.jsx(Km,{open:o,onOpenChange:r,children:n.jsxs(zr,{className:"border sm:max-w-md",style:{background:st,borderColor:ge,color:"hsl(0,0%,100%)"},children:[n.jsxs(jr,{children:[n.jsx(Gr,{className:"font-sans text-lg",children:e("home.modalTitle","Get Your API Key")}),n.jsx(Br,{className:"font-sans text-xs",style:{color:"#FFFFFF"},children:e("home.modalDescription","Enter your email to create a free account instantly.")})]}),n.jsxs("form",{onSubmit:D,className:"mt-4 flex flex-col gap-4",children:[n.jsx(Ht,{type:"email",required:!0,placeholder:e("home.emailPlaceholder","[email protected]"),value:l,onChange:S=>m(S.target.value),className:"border font-sans text-sm",style:{borderColor:ge,background:Ze,color:"#FFFFFF"}}),n.jsx("button",{type:"submit",className:"rounded px-6 py-2.5 font-sans text-sm font-bold uppercase tracking-wider transition-all hover:opacity-90",style:{background:_,color:Ze},children:e("home.createAccount","Create Account â")})]})]})}),n.jsx(_m,{isOpen:s,onClose:()=>d(!1)})]})},wt="infernex_ref_code",In="infernex_ref_source",lo="infernex_ref_fired",co=90;function uo(e,t,a){try{const o=new Date;o.setTime(o.getTime()+a*24*60*60*1e3),document.cookie=`${e}=${encodeURIComponent(t)};expires=${o.toUTCString()};path=/;SameSite=Lax`}catch{}}function sa(e){try{const t=document.cookie.match(new RegExp("(?:^|; )"+e+"=([^;]*)"));return t?decodeURIComponent(t[1]):null}catch{return null}}function Xm(){if(!(typeof window>"u"))try{const e=new URL(window.location.href),t=e.searchParams.get("ref"),a=e.searchParams.get("src");t&&/^[A-Za-z0-9_-]{3,40}$/.test(t)&&(localStorage.setItem(wt,t),uo(wt,t,co)),a&&/^[A-Za-z0-9_-]{2,30}$/.test(a)&&(localStorage.setItem(In,a),uo(In,a,co));const o=localStorage.getItem(wt)||sa(wt);o&&!sessionStorage.getItem(lo)&&(sessionStorage.setItem(lo,"1"),ye.functions.invoke("affiliate-track",{body:{code:o,landingPath:e.pathname,referrer:document.referrer||null}}).catch(()=>{}))}catch{}}function ep(){return typeof window>"u"?{code:null,source:null}:{code:localStorage.getItem(wt)||sa(wt),source:localStorage.getItem(In)||sa(In)}}async function eb(e,t){try{const{code:a,source:o}=ep();if(!a&&!o)return;const r=o==="switch"||o==="openai-switch";await ye.from("signup_attributions").upsert({user_id:e,email:t??null,affiliate_code:a,source:o,switch_bonus_pending:r},{onConflict:"user_id"})}catch(a){console.error("recordAttributionForUser failed",a)}}function tp(){const[e,t]=i.useState([]),[a,o]=i.useState(!0);return i.useEffect(()=>{let s=!0;(async()=>{const{data:{user:l}}=await ye.auth.getUser();if(!l){s&&(t(["guest"]),o(!1));return}const{data:m}=await ye.from("user_roles").select("role").eq("user_id",l.id);s&&(t((m??[]).map(u=>u.role)||["user"]),o(!1))})();const{data:d}=ye.auth.onAuthStateChange(()=>{(async()=>{const{data:{user:l}}=await ye.auth.getUser();if(!l){t(["guest"]);return}const{data:m}=await ye.from("user_roles").select("role").eq("user_id",l.id);t((m??[]).map(u=>u.role)||["user"])})()});return()=>{s=!1,d.subscription.unsubscribe()}},[]),{roles:e,has:(...s)=>s.some(d=>e.includes(d)),loading:a}}const np=["admin","founder","cto","investor","enterprise_client","partner","finance","sales","marketing","developer","user","guest"],Ur={admin:"/founder",founder:"/founder",cto:"/founder",finance:"/finops",sales:"/crm",marketing:"/crm",developer:"/developer-console",investor:"/investor",enterprise_client:"/customer-portal",partner:"/partner-portal",user:"/dashboard",guest:"/dashboard"},ap={admin:"Admin",founder:"Founder OS",cto:"CTO",finance:"Finance",sales:"Sales",marketing:"Marketing",developer:"Developer",investor:"Investor Portal",enterprise_client:"Enterprise Client",partner:"Partner Portal",user:"My Dashboard",guest:"Guest"};function Qr(e){for(const t of np)if(e.includes(t))return t;return"user"}function Vr(e){return Ur[Qr(e)]??"/dashboard"}const tb=Object.freeze(Object.defineProperty({__proto__:null,ROLE_HOME:Ur,ROLE_LABEL:ap,homeForRoles:Vr,primaryRole:Qr},Symbol.toStringTag,{value:"Module"}));function op({roles:e,children:t}){const{roles:a,has:o,loading:r}=tp();return r?n.jsx("div",{className:"min-h-screen flex items-center justify-center bg-background",children:n.jsx(ma,{className:"h-6 w-6 text-primary animate-pulse"})}):o("admin","founder")||o(...e)?n.jsx(n.Fragment,{children:t}):n.jsx(lt,{to:Vr(a),replace:!0})}const rp=i.lazy(()=>f(()=>
1019import("./WorkforceHome-BOB3kghP.js"),__vite__mapDeps([0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30]))),ip=i.lazy(()=>f(()=>import("./MasterOrchestrator-CcrMYKer.js"),__vite__mapDeps([31,1,2,32,4,33,10,11,8,34,35,36,37,38,39,18,22,24,40,41,42,43,25,13,44,45,46,47,48,9,7,17,49,50,51,52,53,28,29,30]))),sp=i.lazy(()=>f(()=>import("./CompanyBuilder-UNPHpAgf.js"),__vite__mapDeps([54,1,2,33,32,4,55,56,28,57,40,58,14,59,60,61,38,25,41,36,22,62,44,53,63,18,64,65,66,67,7,68,42,69,34,16,8,37,70,71,72,21,43,13,73,74,75,76,20,77,78,79,80,17,81,82,24,52,83,29,30]))),lp=i.lazy(()=>f(()=>import("./ReadyMadeCompanies-B_pVX5Iz.js"),__vite__mapDeps([84,1,2,4,8,51,7,40,22,28,29,30]))),Vn=i.lazy(()=>f(()=>import("./PublicSiteRenderer-BA3zcR5x.js"),__vite__mapDeps([85,1,2,4,51,28,29,30]))),cp=i.lazy(()=>f(()=>import("./SuperAdminOS-YaD0Bxzy.js"),__vite__mapDeps([86,1,2,33,32,4,66,38,59,87,56,28,76,88,75,89,16,25,81,55,90,91,24,62,57,22,7,36,72,20,92,27,71,29,30]))),dp=i.lazy(()=>f(()=>import("./RC2InnovationCenter-DBHC5uB8.js"),__vite__mapDeps([93,1,2,33,57,28,32,4,90,94,36,59,20,38,22,62,91,95,29,30]))),up=i.lazy(()=>f(()=>import("./IntegrationCenter-DFRtLC6m.js"),__vite__mapDeps([96,1,2,33,57,28,32,4,94,72,59,89,22,38,62,66,20,39,29,30]))),mp=i.lazy(()=>f(()=>import("./MasterCommandCenter-CHDdbs1Y.js"),__vite__mapDeps([97,1,2,33,57,28,32,4,55,56,94,36,72,24,22,66,7,38,17,59,92,83,40,62,20,29,30]))),pp=i.lazy(()=>f(()=>import("./ApprovalGovernance-I-agcrzb.js"),__vite__mapDeps([98,1,2,33,57,28,32,4,22,59,99,71,66,29,30]))),fp=i.lazy(()=>f(()=>import("./AuthShell-vtmnbNYL.js"),__vite__mapDeps([100,1,2,48,4,28,29,30]))),hp=i.lazy(()=>f(()=>import("./CookieConsent-BxH8aywZ.js"),__vite__mapDeps([101,1,2,4,6,102,28,29,30])).then(e=>({default:e.CookieConsent})));i.lazy(()=>f(()=>import("./AmyHomeGreeting-Bn8UMNLL.js"),__vite__mapDeps([103,1,2,4,48,28,29,30])).then(e=>({default:e.AmyHomeGreeting})));const gp=i.lazy(()=>f(()=>import("./MobileConversionLayer-DTDRhKeu.js"),__vite__mapDeps([104,1,2,4]))),bp=i.lazy(()=>f(()=>import("./ExitIntentPopup-UWTkQUll.js"),__vite__mapDeps([105,1,2,4,106,28,29,30])).then(e=>({default:e.ExitIntentPopup}))),yp=i.lazy(()=>f(()=>import("./FoundersPopup-4qN-7rLg.js"),__vite__mapDeps([107,1,2,108,4,6,28,29,30])).then(e=>({default:e.FoundersPopup}))),vp=i.lazy(()=>f(()=>import("./PlatformCopilot-BXMQLwi5.js"),__vite__mapDeps([109,1,2,4,51,52,28,29,30])).then(e=>({default:e.PlatformCopilot}))),Ip=i.lazy(()=>f(()=>import("./FoundersOffer-DVf4g8hu.js"),__vite__mapDeps([110,1,2,111,28,112,108,4,113,51,7,69,62,17,23,40,6,29,30]))),Ap=i.lazy(()=>f(()=>import("./HowItWorks-CZvM4I85.js"),__vite__mapDeps([114,1,2,4,24,94,73,16,27,72,36,76,28,29,30]))),xp=i.lazy(()=>f(()=>import("./BookDemo-IWnc70zQ.js"),__vite__mapDeps([115,1,2,116,117,28,29,4,30,111,55,32,7,38,118]))),wp=i.lazy(()=>f(()=>import("./Login-BG0C7Z04.js"),__vite__mapDeps([119,1,2,48,4,120,51,121,122,28,29,30]))),Sp=i.lazy(()=>f(()=>import("./Signup-DHjDxGPM.js"),__vite__mapDeps([123,1,2,48,4,6,120,51,124,94,121,122,81,28,29,30]))),kp=i.lazy(()=>f(()=>import("./OAuthConsent-O7JmmFIu.js"),__vite__mapDeps([125,1,2,51,28,29,4,30]))),mo=i.lazy(()=>f(()=>import("./Admin-BBXpUMsr.js"),__vite__mapDeps([126,1,2,48,4,47,57,28,127,124,63,7,21,43,128,79,80,6,129,130,131,132,25,99,133,95,134,111,27,51,118,33,29,135,32,77,56,38,113,13,136,20,82,137,138,139,140,40,141,68,91,142,52,62,19,143,144,145,146,147,39,8,122,148,60,66,149,150,55,151,24,152,153,30,17,41,154,155,70,83]))),Cp=i.lazy(()=>f(()=>import("./Dashboard-DNWToVhZ.js"),__vite__mapDeps([156,1,2,48,4,47,33,57,28,32,127,142,121,122,6,91,40,157,56,158,135,24,63,138,38,53,42,9,21,159,66,43,29,36,20,92,160,46,161,147,95,39,7,155,106,30]))),Pp=i.lazy(()=>f(()=>import("./PaygDashboard-B-Jnwr3w.js"),__vite__mapDeps([162,1,2,47,33,56,28,32,4,129,94,132,60,124,163,22,29,30]))),Tp=i.lazy(()=>f(()=>import("./Privacy-DNecwcrS.js"),__vite__mapDeps([164,1,2,116,117,28,29,4,30,165,94,26,122,7,20,40,16]))),po=i.lazy(()=>f(()=>import("./Cgu-ie82Y6d5.js"),__vite__mapDeps([166,1,2,116,117,28,29,4,30,165,94,16,60,22]))),Ep=i.lazy(()=>f(()=>import("./PaymentSuccess-DIHgWumO.js"),__vite__mapDeps([167,1,2,112,4,6,28,29,30]))),Rp=i.lazy(()=>f(()=>import("./EnterpriseCheckout-CDSsT2oa.js"),__vite__mapDeps([168,1,2,111,28,33,32,4,169,7,12,94,51,142,113,42,29,30]))),Op=i.lazy(()=>f(()=>import("./NotFound-Dp7yXHa2.js"),__vite__mapDeps([170,1,2,4,28,29,30]))),Mp=i.lazy(()=>f(()=>import("./Blog-Cby938Pg.js"),__vite__mapDeps([171,1,2,116,117,28,29,4,30,172,33,32,173,174,47,20,122,137,63,21,44,43,26]))),qp=i.lazy(()=>f(()=>import("./CheapInference-B0qB4dl0.js"),__vite__mapDeps([175,1,2,116,117,28,29,4,30,172,176,174,177,6,77,173,32,94,137,20,38]))),Dp=i.lazy(()=>f(()=>import("./OpenAIAlternative-D8l7kd5v.js"),__vite__mapDeps([178,1,2,116,117,28,29,4,30,172,129,176,174,177,6,77,173,32,94,137,20,179,38,62]))),Lp=i.lazy(()=>f(()=>import("./EconomicLLM-DWOtE3_N.js"),__vite__mapDeps([180,1,2,116,117,28,29,4,30,172,33,176,174,177,6,77,173,32,94,137,20,8,181]))),Fp=i.lazy(()=>f(()=>import("./DynamicBlogPost-DETivjn8.js"),__vite__mapDeps([182,1,2,47,116,117,28,29,4,30,172,127,176,174,177,6,77,173,32,183,94,137,20,122]))),Np=i.lazy(()=>f(()=>import("./AzureVsInfernex-BT5rDawy.js"),__vite__mapDeps([184,1,2,116,117,28,29,4,30,172,176,174,177,6,77,173,32,94,137,20,38]))),_p=i.lazy(()=>f(()=>import("./GoogleCloudAiCosts-Dj3uHtwB.js"),__vite__mapDeps([185,1,2,116,117,28,29,4,30,172,176,174,177,6,77,173,32,94,137,20,38]))),zp=i.lazy(()=>f(()=>import("./InferenceCostPerformance-BiuShDe9.js"),__vite__mapDeps([186,1,2,116,117,28,29,4,30,172,176,174,177,6,77,173,32,94,137,20,38]))),jp=i.lazy(()=>f(()=>import("./Azure70VsInfernex95-D4jnmQxB.js"),__vite__mapDeps([187,1,2,116,117,28,29,4,30,172,176,174,177,6,77,173,32,94,137,20,38]))),Gp=i.lazy(()=>f(()=>import("./SemanticCacheLLM-Cx4_tc_N.js"),__vite__mapDeps([188,1,2,116,117,28,29,4,30,172,176,174,177,6,77,173,32,94,137,20,26,38]))),Bp=i.lazy(()=>f(()=>import("./QuantumInfernex-CZ182VjT.js"),__vite__mapDeps([189,1,2,116,117,28,29,4,30,32,169,172,176,174,177,6,77,94,20,137,24]))),Up=i.lazy(()=>f(()=>import("./CompareOpenAI-CF7cP_i4.js"),__vite__mapDeps([190,1,2,116,117,28,29,4,30,172,32,33,129,94,179,6]))),Qp=i.lazy(()=>f(()=>import("./CompareAWS-BigCrZ-J.js"),__vite__mapDeps([191,1,2,116,117,28,29,4,30,172,32,33,129,94,44,41,179,6]))),Vp=i.lazy(()=>f(()=>import("./CompareGoogle-Dp69bhDM.js"),__vite__mapDeps([192,1,2,116,117,28,29,4,30,172,32,33,129,94,179,6]))),Hp=i.lazy(()=>f(()=>import("./CompareGroq-C3T-S9Wa.js"),__vite__mapDeps([193,1,2,116,117,28,29,4,30,172,32,33,129,94,179,6]))),Wp=i.lazy(()=>f(()=>import("./CompareTogether-CkAwYDKU.js"),__vite__mapDeps([194,1,2,116,117,28,29,4,30,172,32,33,129,94,179,6]))),Kp=i.lazy(()=>f(()=>import("./CompareAnyscale-PhJPFZiE.js"),__vite__mapDeps([195,1,2,116,117,28,29,4,30,172,32,33,129,94,41,179,6]))),$p=i.lazy(()=>f(()=>import("./CompareNvidia-7AfoDFkj.js"),__vite__mapDeps([196,1,2,116,117,28,29,4,30,172,32,33,129,94,17,179,6,63,41]))),Yp=i.lazy(()=>f(()=>import("./CompareHuggingFace-Ctkk7YnS.js"),__vite__mapDeps([197,1,2,116,117,28,29,4,30,172,32,33,129,94,65,179,6]))),Zp=i.lazy(()=>f(()=>import("./Status-BdDuOq_N.js"),__vite__mapDeps([198,1,2,116,117,28,29,4,30,172,32,33,38,66,17,41,40,26,20])));i.lazy(()=>f(()=>import("./Changelog-Bp9pHtJm.js"),__vite__mapDeps([199,1,2,116,117,28,29,4,30,172,32,33,200,24,138])));const Jp=i.lazy(()=>f(()=>import("./Playground-uBdTiCnj.js"),__vite__mapDeps([201,1,2,116,117,28,29,4,30,172,32,33,55,128,79,80,6,48,94,51,143,91,20,63]))),Xp=i.lazy(()=>f(()=>import("./Calculator-BS53-_Ij.js"),__vite__mapDeps([202,1,2,116,117,28,29,4,30,172,32,33,111,128,79,80,6,78,203,158,94,181,65,204,205,135,132]))),ef=i.lazy(()=>f(()=>import("./Onboarding-CTj7SuXy.js"),__vite__mapDeps([206,1,2,33,32,4,207,91,6,160,28,29,30]))),tf=i.lazy(()=>f(()=>import("./ResetPassword-GeBd3TM8.js"),__vite__mapDeps([208,1,2,111,28,29,4,30]))),nf=i.lazy(()=>f(()=>import("./EmailPreview-Dm79NUBP.js"),__vite__mapDeps([209,1,2,128,79,28,80,6,4,29,30]))),af=i.lazy(()=>f(()=>import("./WhiteLabel-sN-pSyFl.js"),__vite__mapDeps([210,1,2,48,4,116,117,28,29,30,111,55,33,32,57,211,24,50,51,52,142,91,7,152,83,43,40,6]))),of=i.lazy(()=>f(()=>import("./Unsubscribe-CCN_WSoQ.js"),__vite__mapDeps([212,1,2,4,51,62,94,28,29,30]))),rf=i.lazy(()=>f(()=>import("./Partners-A8JS8KLL.js"),__vite__mapDeps([213,1,2,116,117,28,29,4,30,214,40,43,138]))),sf=i.lazy(()=>f(()=>import("./PartnerProgram-CiFU_oJw.js"),__vite__mapDeps([215,1,2,33,111,28,55,117,29,4,30,146,45,7,38,36,51]))),lf=i.lazy(()=>f(()=>import("./Quantum-COjH-3FL.js"),__vite__mapDeps([216,1,2,217,4,116,
1019165,94,117,28,29,30,172,218,219,211,32,24,50,51,52,220,205,135,179,169,203,80,6,128,79,130,221,48,222,223,62,8,74,15,21,224]))),cf=i.lazy(()=>f(()=>import("./QuantumVsQuant-Cmd_FThO.js"),__vite__mapDeps([225,1,2,116,117,28,29,4,30,172,219,32,169,222,17,24]))),df=i.lazy(()=>f(()=>import("./QuantumAds-c437Uzdb.js"),__vite__mapDeps([226,1,2,116,117,28,29,4,30,219,32,227,80,169,129,48,63,7,21,24]))),uf=i.lazy(()=>f(()=>import("./QuantumDashboard-D7dpgqEC.js"),__vite__mapDeps([228,1,2,116,117,28,29,4,30,219,172,32,33,48,221,17,224,124]))),mf=i.lazy(()=>f(()=>import("./QuantumDocs-BThuSqa0.js"),__vite__mapDeps([229,1,2,116,117,28,29,4,30,32,169,17]))),pf=i.lazy(()=>f(()=>import("./QuantumOnboarding-B_B4DQKi.js"),__vite__mapDeps([230,1,2,116,117,28,29,4,30,219,172,32,48,221,142,17,24,124]))),ff=i.lazy(()=>f(()=>import("./ReduceAICosts-CLFrLZDo.js"),__vite__mapDeps([231,1,2,116,117,28,29,4,30,172,32,33,63,41,135]))),hf=i.lazy(()=>f(()=>import("./AIInferencePricing-hhxagn1k.js"),__vite__mapDeps([232,1,2,116,117,28,29,4,30,172,32,33,63,224]))),gf=i.lazy(()=>f(()=>import("./WorkforcePricing-AJzYFnWb.js"),__vite__mapDeps([233,1,2,116,117,28,29,4,30,32,33,3,5,22,42,59,6,234,7,8,9]))),bf=i.lazy(()=>f(()=>import("./WorkforcePlanDetail-BwyTg1bp.js"),__vite__mapDeps([235,1,2,116,117,28,29,4,30,32,33,3,5,94,22,42,59,6,234,7,8,9]))),yf=i.lazy(()=>f(()=>import("./HireQuantAlternative-DzZwYjTH.js"),__vite__mapDeps([236,1,2,116,117,28,29,4,30,172,32,33,17,21,63,20,7]))),vf=i.lazy(()=>f(()=>import("./EnterpriseSolutions-DwxnJKyF.js"),__vite__mapDeps([237,1,2,116,117,28,29,4,30,172,32,33,8,224,135,41,40]))),If=i.lazy(()=>f(()=>import("./QuantumAcceleratedInference-BXlICopR.js"),__vite__mapDeps([238,1,2,116,117,28,29,4,30,172,32,33,17,21]))),Af=i.lazy(()=>f(()=>import("./ServerlessAIInference-DVLV4wS-.js"),__vite__mapDeps([239,1,2,116,117,28,29,4,30,32,172,44,135,20,41]))),xf=i.lazy(()=>f(()=>import("./OpenAIMigrationGuide-Q2ABqRk5.js"),__vite__mapDeps([240,1,2,116,117,28,29,4,30,32,169,172,20,135,157]))),wf=i.lazy(()=>f(()=>import("./I18nAudit-DYvEui24.js"),__vite__mapDeps([241,1,2,116,117,28,29,4,30,33,32,24,51,38,132,113]))),Sf=i.lazy(()=>f(()=>import("./SiteAudit-D7wXTf5P.js"),__vite__mapDeps([242,1,2,48,4,33,32,51,124,143,66,38,113,28,29,30]))),kf=i.lazy(()=>f(()=>import("./GscDashboard-B9ctj54f.js"),__vite__mapDeps([243,1,2,48,4,33,32,51,124,244,122,21,68,28,29,30]))),Cf=i.lazy(()=>f(()=>import("./WorkforceLaunch-B83mZ01a.js"),__vite__mapDeps([245,1,2,48,4,33,32,52,22,61,51,7,8,38,62,20,28,29,30]))),Pf=i.lazy(()=>f(()=>import("./LeadsAnalytics-WUJRh-o3.js"),__vite__mapDeps([246,1,2,48,4,33,32,128,79,28,80,6,129,59,7,21,63,88,82,29,30]))),Tf=i.lazy(()=>f(()=>import("./TranslationGenerator-d33v4Zug.js"),__vite__mapDeps([247,1,2,116,117,28,29,4,30,33,32,56,24,38,113,51,132]))),Ef=i.lazy(()=>f(()=>import("./Photonique-Cq16dDFI.js"),__vite__mapDeps([248,1,2,116,165,4,94,117,28,29,30,32,172,48,158,205,135,24,204,91]))),Rf=i.lazy(()=>f(()=>import("./HardwareVsApiKeys-zM1I1iEo.js"),__vite__mapDeps([249,1,2,116,117,28,29,4,30,172,176,174,177,6,77,94,137,20,41,38,17,63]))),Of=i.lazy(()=>f(()=>import("./TSUPBit-Cu3uRXJV.js"),__vite__mapDeps([250,1,2,116,165,4,94,117,28,29,30,217,220,32,172,205,135,24,179,124,143,53,38]))),fo=i.lazy(()=>f(()=>import("./UltimateSupremacy-YrssIwoN.js"),__vite__mapDeps([251,1,2,217,4,116,165,94,117,28,29,30,32,172,111,220,205,135,24,179,48,51,66]))),Mf=i.lazy(()=>f(()=>import("./CompareAnthropic-3tVCGkVg.js"),__vite__mapDeps([252,1,2,116,117,28,29,4,30,172,32,33,129,94,6,179]))),qf=i.lazy(()=>f(()=>import("./CompareMistral-COcVpi5T.js"),__vite__mapDeps([253,1,2,116,117,28,29,4,30,172,32,33,129,94,6,179]))),Df=i.lazy(()=>f(()=>import("./SwitchFromOpenAI-CkDFx3Ik.js"),__vite__mapDeps([254,1,2,116,117,28,29,4,30,172,32,94,137,20]))),Lf=i.lazy(()=>f(()=>import("./Models-BuJ2MFCI.js"),__vite__mapDeps([255,1,2,116,117,28,29,4,30,172,32,129,25,6]))),Ff=i.lazy(()=>f(()=>import("./GEO-CCn5FnzL.js"),__vite__mapDeps([256,1,2,116,172,4,220,32,205,135,24,179,158,6,234,28,29,30]))),Nf=i.lazy(()=>f(()=>import("./GeoDashboard-bubhzVMM.js"),__vite__mapDeps([257,1,2,48,4,47,116,117,28,29,30,55]))),_f=i.lazy(()=>f(()=>import("./MiroFish-CrTk1eZu.js"),__vite__mapDeps([258,1,2,116,165,4,94,32,172,48,57,28,259,211,24,50,51,52,220,205,135,179,6,91,21,40,60,22,44,7,16,29,30]))),zf=i.lazy(()=>f(()=>import("./MiroFishInference-hAYTZLi0.js"),__vite__mapDeps([260,1,2,116,117,28,29,4,30,172,176,174,177,6,77,173,32,94,137,20])
1019)),jf=i.lazy(()=>f(()=>import("./MiroFishSimulation-ZHe87aPg.js"),__vite__mapDeps([261,1,2,116,117,28,29,4,30,172]))),Gf=i.lazy(()=>f(()=>import("./MiroFishFAQ-BWDypKSx.js"),__vite__mapDeps([262,1,2,116,117,28,29,4,30,172]))),Bf=i.lazy(()=>f(()=>import("./MiroFishDocs-D1X-eRr_.js"),__vite__mapDeps([263,1,2,116,117,28,29,4,30,172]))),Uf=i.lazy(()=>f(()=>import("./IAO-AHRPhwlm.js"),__vite__mapDeps([264,1,2,116,117,28,29,4,30,172,219,32,33,48,220,205,135,24,179,51,22,19,23,6,204]))),Qf=i.lazy(()=>f(()=>import("./IAOChat-DPFZkh-U.js"),__vite__mapDeps([265,1,2,116,117,28,29,4,30,48,46,47,150,55,151,79,24,51,19,60,52,6,91,152,141,99,153]))),Vf=i.lazy(()=>f(()=>import("./IAODemo-BWT0Ywht.js"),__vite__mapDeps([266,1,2,55,151,28,4,79,117,29,30,24,19,6,91,51,60,52]))),Hf=i.lazy(()=>f(()=>import("./Platform-xVIMJNoz.js"),__vite__mapDeps([267,1,2,4,17,135,204,65,24,268,138,28,29,30]))),Wf=i.lazy(()=>f(()=>import("./WidgetIAOChat-CsYM5xlz.js"),__vite__mapDeps([269,1,2,4,24,19,51,52,6,91,28,29,30]))),Kf=i.lazy(()=>f(()=>import("./IAOWidgetInstall-CncRLTc6.js"),__vite__mapDeps([270,1,2,116,117,28,29,4,30,48,33,32,51,40,9,8,142,23,6,152,91,99,35]))),$f=i.lazy(()=>f(()=>import("./AuditResult-DPxhqNHZ.js"),__vite__mapDeps([271,1,2,4,28,29,30]))),Yf=i.lazy(()=>f(()=>import("./AuditDashboard-B7GzuI8c.js"),__vite__mapDeps([272,1,2,4,28,29,30]))),Zf=i.lazy(()=>f(()=>import("./QuadFusionAGI-BQ0vh3s2.js"),__vite__mapDeps([273,1,2,116,117,28,29,4,30,172,173,32,174,176,177,6,77,94,137,20,24,17,204]))),Jf=i.lazy(()=>f(()=>import("./Supremacy-DePqTTw-.js"),__vite__mapDeps([274,1,2,117,28,29,4,30,220,32,172,205,135,24,179,40,160,132]))),Xf=i.lazy(()=>f(()=>import("./PartnerPage-BEwm8aJM.js"),__vite__mapDeps([275,1,2,4,28,29,30]))),eh=i.lazy(()=>f(()=>import("./TrustCenter-C1r8rJpG.js"),__vite__mapDeps([276,1,2,116,4,223,40,28,29,30]))),th=i.lazy(()=>f(()=>import("./Unsubscribe-CVh-kklx.js"),__vite__mapDeps([277,1,2,4,51,38,62,28,29,30]))),nh=i.lazy(()=>f(()=>import("./Themis-Dz1vj9lh.js"),__vite__mapDeps([278,1,2,116,220,32,4,172,205,135,24,179,279,22,40,223,280,20,94,16,25,6,28,29,30]))),ah=i.lazy(()=>f(()=>import("./ThemisDashboard-Cg5JtjPq.js"),__vite__mapDeps([281,1,2,48,4,51,94,16,140,280,60,22,279,99,25,28,29,30]))),ho=i.lazy(()=>f(()=>import("./ThemisVerify-BG_PwPoM.js"),__vite__mapDeps([282,1,2,4,94,16,22,51,25,62,280,20,28,29,30]))),un=i.lazy(()=>f(()=>import("./BenchmarkVerify-CY0wNvzf.js"),__vite__mapDeps([283,1,2,4,94,22,160,179,51,62,6,91,280,20,28,29,30]))),oh=i.lazy(()=>f(()=>import("./Switch-D1ezqIr9.js"),__vite__mapDeps([284,1,2,117,28,29,4,30,22,140,51,60,135]))),rh=i.lazy(()=>f(()=>import("./VsOpenAILive-SVBCEvpU.js"),__vite__mapDeps([285,1,2,117,28,29,4,30,51,179,177]))),ih=i.lazy(()=>f(()=>import("./Affiliates-CqSdlszC.js"),__vite__mapDeps([286,1,2,117,28,29,4,30,63,91,244,7]))),sh=i.lazy(()=>f(()=>import("./Sdk-DUII8XDd.js"),__vite__mapDeps([287,1,2,117,28,29,4,30,288,91]))),go=i.lazy(()=>f(()=>import("./Developers-X0pxOUor.js"),__vite__mapDeps([289,1,2,117,28,29,4,30,33,57,32,23,8,288,36,45,142,154,60,160,91,7,6]))),lh=i.lazy(()=>f(()=>import("./Testimonials-CV95L0yx.js"),__vite__mapDeps([290,1,2,4,268,138,38,28,29,30]))),ch=i.lazy(()=>f(()=>import("./Support-4P5X6JVt.js"),__vite__mapDeps([291,1,2,4,45,7,38,113,34,118,28,29,30]))),dh=i.lazy(()=>f(()=>import("./Founder-CoS0VsAd.js"),__vite__mapDeps([292,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,295,33,44,29,30]))),uh=i.lazy(()=>f(()=>import("./Investor-DS-Yn7lB.js"),__vite__mapDeps([296,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,295,33,56,297,223,29,30]))),mh=i.lazy(()=>f(()=>import("./Documents-DGj_yi6o.js"),__vite__mapDeps([298,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,51,132,29,30]))),ph=i.lazy(()=>f(()=>import("./CRM-CNBrgkKN.js"),__vite__mapDeps([299,1,2,47,131,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,301,33,25,302,111,57,152,29,30]))),fh=i.lazy(()=>f(()=>import("./EnterpriseAccounts-CYgLAGdD.js"),__vite__mapDeps([303,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,301,33,25,29,30]))),hh=i.lazy(()=>f(()=>import("./PartnerCenter-BFAbeCNU.js"),__vite__mapDeps([304,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,301,33,25,295,128,79,80,132,91,51,305,29,30]))),gh=i.lazy(()=>f(()=>import("./MarketplaceCenter-CpVHnHDT.js"),__vite__mapDeps([306,1,2,47,131,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,33,56,124,307,66,20,62,29,30]))),bh=i.lazy(()=>f(()=>import("./FinOps-B3_6avt3.js"),__vite__mapDeps([308,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,301,33,25,295,29,30]))),yh=i.lazy(()=>f(()=>import("./BizDev-C349iLg4.js"),__vite__mapDeps([309,1,2,47,131,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,301,33,25,302,295,111,128,79,80,57,152,29,30]))),vh=i.lazy(()=>f(()=>import("./DeveloperConsole-BndKbfMe.js"),__vite__mapDeps([310,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,57,160,154,143,91,288,29,30]))),Ih=i.lazy(()=>f(()=>import("./DataRoom-DFRRwqYf.js"),__vite__mapDeps([311,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,25,51,20,132,29,30]))),Ah=i.lazy(()=>f(()=>import("./PitchDeck-C8YQwpOa.js"),__vite__mapDeps([312,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,297,313,314,133,51,307,143,29,30]))),xh=i.lazy(()=>f(()=>import("./DueDiligence-DZTiFAEg.js"),__vite__mapDeps([315,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,56,297,313,51,307,29,30]))),wh=i.lazy(()=>f(()=>import("./PredictiveAnalytics-CWdnnxad.js"),__vite__mapDeps([316,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,313,29,50,30]))),Sh=i.lazy(()=>f(()=>import("./Copilot-DMCf8rAS.js"),__vite__mapDeps([317,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,149,150,55,151,79,51,52,91,152,141,99,48,153,29,30]))),kh=i.lazy(()=>f(()=>
1019import("./AiSupervisor-DMMVqVxC.js"),__vite__mapDeps([318,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,56,297,313,64,29,30]))),Ch=i.lazy(()=>f(()=>import("./RoiEngine-BDWhuo3K.js"),__vite__mapDeps([319,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,313,29,50,30]))),Ph=i.lazy(()=>f(()=>import("./AutomationCenter-De59C7hN.js"),__vite__mapDeps([320,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,20,51,143,152,62,148,29,30]))),Th=i.lazy(()=>f(()=>import("./Directories-Bj0_cXiB.js"),__vite__mapDeps([321,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,203,80,56,51,124,113,20,307,139,52,143,91,29,30]))),Eh=i.lazy(()=>f(()=>import("./PartnershipRoadmap-DDEkYiPx.js"),__vite__mapDeps([322,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,128,79,80,59,63,51,307,77,29,30]))),Rh=i.lazy(()=>f(()=>import("./KnowledgeCenter-DbjzGG_2.js"),__vite__mapDeps([323,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,313,25,50,29,30])));i.lazy(()=>f(()=>import("./StubPage-cxiC0Fok.js"),__vite__mapDeps([324,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,29,30])).then(e=>({default:e.EnterpriseSettings})));const Oh=i.lazy(()=>f(()=>import("./DeveloperConsole-BndKbfMe.js"),__vite__mapDeps([310,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,57,160,154,143,91,288,29,30]))),Mh=i.lazy(()=>f(()=>import("./SecurityCenter-DHH_0sXJ.js"),__vite__mapDeps([325,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,305,29,30]))),qh=i.lazy(()=>f(()=>import("./GlobalOptimizer-CDp4Fvga.js"),__vite__mapDeps([326,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,297,33,313,327,124,29,30]))),Dh=i.lazy(()=>f(()=>import("./ExecutiveInsights-DBYz0A1D.js"),__vite__mapDeps([328,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,313,33,327,132,50,66,29,30]))),Lh=i.lazy(()=>f(()=>import("./UserAnalytics-jsJBM76b.js"),__vite__mapDeps([329,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,33,25,29,30]))),Fh=i.lazy(()=>f(()=>import("./WarRoomYC-BV1mnJbJ.js"),__vite__mapDeps([330,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,33,62,135,66,63,29,30]))),Nh=i.lazy(()=>f(()=>import("./AiWorkforce-DYv148JU.js"),__vite__mapDeps([331,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,55,128,79,80,57,152,52,145,99,29,30]))),_h=i.lazy(()=>f(()=>import("./AiWorkforceOnboarding-F0DLQ7Kn.js"),__vite__mapDeps([332,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,72,141,29,30]))),zh=i.lazy(()=>f(()=>import("./AiWorkforceChat-CbVUZ21P.js"),__vite__mapDeps([333,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,55,94,52,145,29,30]))),jh=i.lazy(()=>f(()=>import("./WorkforceDashboard-vi1DxFfr.js"),__vite__mapDeps([334,1,2,33,32,4,127,335,207,24,59,91,38,62,7,19,20,63,21,29,67,12,13,181,16,25,43,28,30]))),Gh=i.lazy(()=>f(()=>import("./WorkforceConnectors-CPrnlNsG.js"),__vite__mapDeps([336,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,11,51,
1019113,141,58,29,30]))),Bh=i.lazy(()=>f(()=>import("./WorkforceRouter-BEhN72-I.js"),__vite__mapDeps([337,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,129,135,29,30]))),Uh=i.lazy(()=>f(()=>import("./WorkforceSafety-DX7uHaqM.js"),__vite__mapDeps([338,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,66,62,29,30]))),Qh=i.lazy(()=>f(()=>import("./GlobalIntelligenceCenter-uk-Zke9p.js"),__vite__mapDeps([339,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,51,148,29,30]))),Vh=i.lazy(()=>f(()=>import("./PartnerIntelligenceCenter-CDOx3I2q.js"),__vite__mapDeps([340,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,301,25,56,341,51,49,66,29,30]))),Hh=i.lazy(()=>f(()=>import("./FinOpsIntelligenceCenter-CwUu_Y46.js"),__vite__mapDeps([342,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,56,341,51,49,66,44,135,63,29,30]))),Wh=i.lazy(()=>f(()=>import("./ComplianceCenter-Cen3h48p.js"),__vite__mapDeps([343,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,56,341,51,49,66,314,132,62,29,30]))),Kh=i.lazy(()=>f(()=>import("./ExecutiveCommandCenter-CJdSM2dy.js"),__vite__mapDeps([344,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,33,55,56,341,51,49,66,91,132,50,29,30]))),$h=i.lazy(()=>f(()=>import("./QualityGate-BXj2QXOs.js"),__vite__mapDeps([345,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,56,314,51,132,66,62,29,30]))),Yh=i.lazy(()=>f(()=>import("./KnowledgeGraph-BF0HQQ6D.js"),__vite__mapDeps([346,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,33,25,29,30]))),Zh=i.lazy(()=>f(()=>import("./PartnerPortal-XJ0f_wF_.js"),__vite__mapDeps([347,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,132,29,30]))),Jh=i.lazy(()=>f(()=>import("./CustomerPortal-C7IOr_Iv.js"),__vite__mapDeps([348,1,2,47,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,295,33,313,29,30]))),Xh=i.lazy(()=>f(()=>import("./EnterpriseSettings-gbNle8-d.js"),__vite__mapDeps([349,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,33,227,80,111,40,29,30]))),eg=i.lazy(()=>f(()=>import("./SystemVerification-DzgCw_Vs.js"),__vite__mapDeps([350,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,313,297,33,29,30]))),tg=i.lazy(()=>f(()=>import("./ReferralsDashboard-DOFlTxTF.js"),__vite__mapDeps([351,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,77,91,63,29,30]))),ng=i.lazy(()=>f(()=>import("./EmailMonitor-DSeWNMg0.js"),__vite__mapDeps([352,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,128,79,80,124,132,144,66,62,99,159,29,30]))),ag=i.lazy(()=>f(()=>import("./ProspectTargets-BKn-muWC.js"),__vite__mapDeps([353,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,227,80,143,152,40,99,29,30]))),og=i.lazy(()=>f(()=>import("./OutreachDashboard-D8D8aHio.js"),__vite__mapDeps([354,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,124,40,29,30]))),rg=i.lazy(()=>f(()=>import("./AuditLog-B2xLRYzM.js"),__vite__mapDeps([355,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,29,30]))),ig=i.lazy(()=>f(()=>import("./AiGovernance-Ce3GpJdR.js"),__vite__mapDeps([356,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,300,94,33,57,55,227,80,111,152,99,66,29,30]))),sg=i.lazy(()=>f(()=>import("./AiPerformance-BhOIkF_x.js"),__vite__mapDeps([357,1,2,33,32,4,57,28,127,129,335,207,293,163,134,161,147,95,6,39,7,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,132,20,63,29,30]))),bo=i.lazy(()=>f(()=>import("./WorkforceMarketplace-B6jDNEw4.js"),__vite__mapDeps([358,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,141,25,72,200,12,13,16,15,29,30]))),lg=i.lazy(()=>f(()=>import("./ConnectorHub-CSMei4X4.js"),__vite__mapDeps([359,1,2,11,360,33,32,4,361,28,29,30]))),cg=i.lazy(()=>f(()=>import("./ConnectorDetail-CjmkSnze.js"),__vite__mapDeps([362,1,2,11,10,33,32,4,361,28,29,30]))),dg=i.lazy(()=>f(()=>import("./EmployeeHub-DXNSMqdI.js"),__vite__mapDeps([363,1,2,10,360,11,33,32,4,361,28,29,30]))),ug=i.lazy(()=>f(()=>import("./EmployeeDetail-WqWsm6v-.js"),__vite__mapDeps([364,1,2,10,11,33,32,4,361,28,29,30]))),mg=i.lazy(()=>f(()=>import("./FirecrawlHealth-DGqRCjzU.js"),__vite__mapDeps([365,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,124,81,51,92,62,66,87,29,30]))),pg=i.lazy(()=>f(()=>import("./DeliverabilityDashboard-CJRWNJdk.js"),__vite__mapDeps([366,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,128,79,80,29,30]))),fg=i.lazy(()=>f(()=>import("./CampaignPreview-B1Oh3jRN.js"),__vite__mapDeps([367,1,2,293,28,4,163,134,161,147,95,6,39,32,7,207,22,19,53,18,17,9,21,8,67,294,76,23,26,181,37,75,36,45,24,34,60,38,68,102,33,55,128,79,80,29,30]))),N=(e,t)=>n.jsx(op,{roles:e,children:t}),hg=new sc,gg=()=>n.jsx("div",{className:"flex min-h-screen items-center justify-center bg-background",children:n.jsx("div",{className:"h-10 w-10 animate-spin rounded-full border-2 border-primary border-t-transparent"})}),bg={"moltcruse.com":"moltcruse","www.moltcruse.com":"moltcruse"},yg=()=>{const[e,t]=i.useState(!1),a=typeof window>"u"?null:bg[window.location.hostname.toLowerCase()]??null;return i.useEffect(()=>{Xm()},[]),i.useEffect(()=>{let o,r;const s=()=>t(!0);return"requestIdleCallback"in window?r=window.requestIdleCallback(s,{timeout:3500}):o=globalThis.setTimeout(s,2500),()=>{r!==void 0&&"cancelIdleCallback"in window&&window.cancelIdleCallback(r),o!==void 0&&globalThis.clearTimeout(o)}},[]),n.jsx(zc,{children:n.jsx(Mr,{children:n.jsx(lc,{client:hg,children:n.jsx(rl,{attribute:"class",defaultTheme:"dark",enableSystem:!0,disableTransitionOnChange:!1,children:n.jsxs(Nl,{children:[n.jsx(tl,{}),n.jsx(br,{}),n.jsx(Im,{children:n.jsxs(ci,{children:[n.jsx(_c,{}),n.jsx(i.Suspense,{fallback:n.jsx(gg,{}),children:a?n.jsx(Ta,{children:n.jsx(p,{path:"*",element:n.jsx(Vn,{forcedSlug:
1019a})})}):n.jsxs(Ta,{children:[n.jsx(p,{path:"/",element:n.jsx(rp,{})}),n.jsx(p,{path:"/legacy-home",element:n.jsx(so,{})}),n.jsx(p,{path:"/index",element:n.jsx(so,{})}),n.jsx(p,{path:"/workforce-home",element:n.jsx(lt,{to:"/",replace:!0})}),n.jsx(p,{path:"/pricing",element:n.jsx(lt,{to:"/ai-inference-pricing",replace:!0})}),n.jsx(p,{path:"/docs",element:n.jsx(lt,{to:"/docs/openai-migration-guide",replace:!0})}),n.jsx(p,{path:"/openai-migration-guide",element:n.jsx(lt,{to:"/docs/openai-migration-guide",replace:!0})}),n.jsx(p,{path:"/build",element:n.jsx(lt,{to:"/onboarding",replace:!0})}),n.jsx(p,{path:"/billing",element:n.jsx(lt,{to:"/billing/payg",replace:!0})}),n.jsx(p,{path:"/.lovable/oauth/consent",element:n.jsx(kp,{})}),n.jsx(p,{path:"/founders-offer",element:n.jsx(Ip,{})}),n.jsx(p,{path:"/how-it-works",element:n.jsx(Ap,{})}),n.jsx(p,{path:"/ready-made-companies",element:n.jsx(lp,{})}),n.jsx(p,{path:"/sites/:slug",element:n.jsx(Vn,{})}),n.jsx(p,{path:"/sites/:slug/*",element:n.jsx(Vn,{})}),n.jsx(p,{path:"/book-demo",element:n.jsx(xp,{})}),n.jsx(p,{path:"/marketplace/connectors",element:n.jsx(lg,{})}),n.jsx(p,{path:"/marketplace/connector/:id",element:n.jsx(cg,{})}),n.jsx(p,{path:"/marketplace/employees",element:n.jsx(dg,{})}),n.jsx(p,{path:"/marketplace/employee/:id",element:n.jsx(ug,{})}),n.jsxs(p,{element:n.jsx(fp,{}),children:[n.jsx(p,{path:"/login",element:n.jsx(wp,{})}),n.jsx(p,{path:"/signup",element:n.jsx(Sp,{})}),n.jsx(p,{path:"/reset-password",element:n.jsx(tf,{})}),n.jsx(p,{path:"/orchestrator",element:n.jsx(ip,{})}),n.jsx(p,{path:"/company-builder",element:n.jsx(sp,{})}),n.jsx(p,{path:"/super-admin",element:n.jsx(cp,{})}),n.jsx(p,{path:"/super-admin/rc2",element:n.jsx(dp,{})}),n.jsx(p,{path:"/super-admin/integrations",element:n.jsx(up,{})}),n.jsx(p,{path:"/super-admin/command-center",element:n.jsx(mp,{})}),n.jsx(p,{path:"/super-admin/approvals",element:n.jsx(pp,{})}),n.jsx(p,{path:"/admin",element:n.jsx(mo,{})}),n.jsx(p,{path:"/admin/founders",element:n.jsx(mo,{})}),n.jsx(p,{path:"/dashboard",element:n.jsx(Cp,{})}),n.jsx(p,{path:"/billing/payg",element:n.jsx(Pp,{})}),n.jsx(p,{path:"/privacy",element:n.jsx(Tp,{})}),n.jsx(p,{path:"/cgu",element:n.jsx(po,{})}),n.jsx(p,{path:"/terms",element:n.jsx(po,{})}),n.jsx(p,{path:"/payment-success",element:n.jsx(Ep,{})}),n.jsx(p,{path:"/enterprise-checkout",element:n.jsx(Rp,{})}),n.jsx(p,{path:"/blog",element:n.jsx(Mp,{})}),n.jsx(p,{path:"/testimonials",element:n.jsx(lh,{})}),n.jsx(p,{path:"/support",element:n.jsx(ch,{})}),n.jsx(p,{path:"/blog/inference-ai-pas-cher",element:n.jsx(qp,{})}),n.jsx(p,{path:"/blog/alternative-openai-2026",element:n.jsx(Dp,{})}),n.jsx(p,{path:"/blog/llm-economique-entreprise",element:n.jsx(Lp,{})}),n.jsx(p,{path:"/blog/azure-ai-costs-vs-infernex",element:n.jsx(Np,{})}),n.jsx(p,{path:"/blog/google-cloud-ai-costs-vs-infernex",element:n.jsx(_p,{})}),n.jsx(p,{path:"/blog/ai-inference-cost-performance-benchmark-2026",element:n.jsx(zp,{})}),n.jsx(p,{path:"/blog/cut-azure-ai-costs-70-vs-infernex-95",element:n.jsx(jp,{})}),n.jsx(p,{path:"/blog/optimise-llm-costs-semantic-cache",element:n.jsx(Gp,{})}),n.jsx(p,{path:"/blog/quantum-infernex",element:n.jsx(Bp,{})}),n.jsx(p,{path:"/blog/hardware-vs-api-keys-infernex",element:n.jsx(Rf,{})}),n.jsx(p,{path:"/blog/switch-from-openai-save-money",element:n.jsx(Df,{})}),n.jsx(p,{path:"/blog/quad-fusion-agi-breakthrough",element:n.jsx(Zf,{})}),n.jsx(p,{path:"/blog/:slug",element:n.jsx(Fp,{})}),n.jsx(p,{path:"/compare/openai",element:n.jsx(Up,{})}),n.jsx(p,{path:"/compare/aws-bedrock",element:n.jsx(Qp,{})}),n.jsx(p,{path:"/compare/google-vertex",element:n.jsx(Vp,{})}),n.jsx(p,{path:"/compare/groq",element:n.jsx(Hp,{})}),n.jsx(p,{path:"/compare/together",element:n.jsx(Wp,{})}),n.jsx(p,{path:"/compare/anyscale",element:n.jsx(Kp,{})}),n.jsx(p,{path:"/compare/nvidia",element:n.jsx($p,{})}),n.jsx(p,{path:"/compare/huggingface",element:n.jsx(Yp,{})}),n.jsx(p,{path:"/compare/anthropic",element:n.jsx(Mf,{})}),n.jsx(p,{path:"/compare/mistral",element:n.jsx(qf,{})}),n.jsx(p,{path:"/status",element:n.jsx(Zp,{})}),n.jsx(p,{path:"/playground",element:n.jsx(Jp,{})}),n.jsx(p,{path:"/calculator",element:n.jsx(Xp,{})}),n.jsx(p,{path:"/onboarding",element:n.jsx(ef,{})}),n.jsx(p,{path:"/white-label",element:n.jsx(af,{})}),n.jsx(p,{path:"/partners",element:n.jsx(rf,{})}),n.jsx(p,{path:"/partner-program",element:n.jsx(sf,{})}),n.jsx(p,{path:"/email-preview",element:n.jsx(nf,{})}),n.jsx(p,{path:"/unsubscribe",element:n.jsx(of,{})}),n.jsx(p,{path:"/quantum",element:n.jsx(lf,{})}),n.jsx(p,{path:"/quantum-vs-quant",element:n.jsx(cf,{})}),n.jsx(p,{path:"/quantum-ads",element:n.jsx(df,{})}),n.jsx(p,{path:"/quantum/dashboard",element:n.jsx(uf,{})}),n.jsx(p,{path:"/quantum/docs",element:n.jsx(mf,{})}),n.jsx(p,{path:"/quantum/onboarding",element:n.jsx(pf,{})}),n.jsx(p,{path:"/reduce-ai-costs",element:n.jsx(ff,{})}),n.jsx(p,{path:"/ai-inference-pricing",element:n.jsx(hf,{})}),n.jsx(p,{path:"/workforce-pricing",element:n.jsx(gf,{})}),n.jsx(p,{path:"/workforce-pricing/:tier",element:n.jsx(bf,{})}),n.jsx(p,{path:"/hire-quant-alternative",element:n.jsx(yf,{})}),n.jsx(p,{path:"/enterprise-solutions",element:n.jsx(vf,{})}),n.jsx(p,{path:"/quantum-accelerated-inference",element:n.jsx(If,{})}),n.jsx(p,{path:"/serverless-ai-inference",element:n.jsx(Af,{})}),n.jsx(p,{path:"/docs/openai-migration-guide",element:n.jsx(xf,{})}),n.jsx(p,{path:"/admin/i18n-audit",element:n.jsx(wf,{})}),n.jsx(p,{path:"/admin/translation-generator",element:n.jsx(Tf,{})}),n.jsx(p,{path:"/admin/site-audit",element:n.jsx(Sf,{})}),n.jsx(p,{path:"/admin/gsc",element:n.jsx(kf,{})}),n.jsx(p,{path:"/admin/workforce-launch",element:n.jsx(Cf,{})}),n.jsx(p,{path:"/admin/leads-analytics",element:n.jsx(Pf,{})}),n.jsx(p,{path:"/photonique",element:n.jsx(Ef,{})}),n.jsx(p,{path:"/models",element:n.jsx(Lf,{})}),n.jsx(p,{path:"/tsu-pbit",element:n.jsx(Of,{})}),n.jsx(p,{path:"/ultimate-inference",element:n.jsx(fo,{})}),n.jsx(p,{path:"/ultimate-supremacy",element:n.jsx(fo,{})}),n.jsx(p,{path:"/geo",element:n.jsx(Ff,{})}),n.jsx(p,{path:"/geo/dashboard",element:n.jsx(Nf,{})}),n.jsx(p,{path:"/mirofish",element:n.jsx(_f,{})}),n.jsx(p,{path:"/blog/mirofish-infernex-cheapest-inference",element:n.jsx(zf,{})}),n.jsx(p,{path:"/blog/mirofish-moteur-ia-simulation-prediction",element:n.jsx(jf,{})}),n.jsx(p,{path:"/faq/mirofish",element:n.jsx(Gf,{})}),n.jsx(p,{path:"/docs/mirofish-integration",element:n.jsx(Bf,{})}),n.jsx(p,{path:"/iao",element:n.jsx(Uf,{})}),n.jsx(p,{path:"/iao/chat",element:n.jsx(Qf,{})}),n.jsx(p,{path:"/iao/demo",element:n.jsx(Vf,{})}),n.jsx(p,{path:"/platform",element:n.jsx(Hf,{})}),n.jsx(p,{path:"/iao/widget",element:n.jsx(Kf,{})}),n.jsx(p,{path:"/audit-result",element:n.jsx($f,{})}),n.jsx(p,{path:"/audit-dashboard",element:n.jsx(Yf,{})}),n.jsx(p,{path:"/supremacy",element:n.jsx(Jf,{})}),n.jsx(p,{path:"/cabinets/:slug",element:n.jsx(Xf,{})}),n.jsx(p,{path:"/themis/trust-center",element:n.jsx(eh,{})}),n.jsx(p,{path:"/themis/unsubscribe",element:n.jsx(th,{})}),n.jsx(p,{path:"/themis/dashboard",element:n.jsx(ah,{})}),n.jsx(p,{path:"/themis",element:n.jsx(nh,{})}),n.jsx(p,{path:"/switch",element:n.jsx(oh,{})}),n.jsx(p,{path:"/vs-openai-live",element:n.jsx(rh,{})}),n.jsx(p,{path:"/affiliates",element:n.jsx(ih,{})}),n.jsx(p,{path:"/sdk",element:n.jsx(sh,{})}),n.jsx(p,{path:"/developers",element:n.jsx(go,{})}),n.jsx(p,{path:"/developer-hub",element:n.jsx(go,{})}),n.jsx(p,{path:"/founder",element:N(["founder","admin","cto"],n.jsx(dh,{}))}),n.jsx(p,{path:"/investor",element:N(["investor","founder","admin"],n.jsx(uh,{}))}),n.jsx(p,{path:"/documents",element:n.jsx(mh,{})}),n.jsx(p,{path:"/crm",element:N(["sales","marketing","founder","admin"],n.jsx(ph,{}))}
1019),n.jsx(p,{path:"/enterprise-accounts",element:N(["sales","founder","admin"],n.jsx(fh,{}))}),n.jsx(p,{path:"/partner-center",element:N(["partner","sales","founder","admin"],n.jsx(hh,{}))}),n.jsx(p,{path:"/marketplace-center",element:N(["marketing","founder","admin"],n.jsx(gh,{}))}),n.jsx(p,{path:"/finops",element:N(["finance","founder","admin"],n.jsx(bh,{}))}),n.jsx(p,{path:"/business-development",element:N(["sales","founder","admin"],n.jsx(yh,{}))}),n.jsx(p,{path:"/developer-console",element:N(["developer","cto","founder","admin"],n.jsx(vh,{}))}),n.jsx(p,{path:"/data-room",element:N(["investor","founder","admin"],n.jsx(Ih,{}))}),n.jsx(p,{path:"/pitch-deck",element:N(["investor","founder","admin"],n.jsx(Ah,{}))}),n.jsx(p,{path:"/due-diligence",element:N(["investor","founder","admin"],n.jsx(xh,{}))}),n.jsx(p,{path:"/analytics",element:n.jsx(wh,{})}),n.jsx(p,{path:"/copilot",element:n.jsx(Sh,{})}),n.jsx(p,{path:"/ai-supervisor",element:N(["cto","founder","admin"],n.jsx(kh,{}))}),n.jsx(p,{path:"/roi-engine",element:N(["finance","investor","founder","admin"],n.jsx(Ch,{}))}),n.jsx(p,{path:"/automation",element:N(["cto","founder","admin"],n.jsx(Ph,{}))}),n.jsx(p,{path:"/directories",element:N(["marketing","founder","admin"],n.jsx(Th,{}))}),n.jsx(p,{path:"/partnership-roadmap",element:N(["founder","admin"],n.jsx(Eh,{}))}),n.jsx(p,{path:"/knowledge",element:n.jsx(Rh,{})}),n.jsx(p,{path:"/settings",element:n.jsx(Xh,{})}),n.jsx(p,{path:"/developer-console-v2",element:N(["developer","cto","founder","admin"],n.jsx(Oh,{}))}),n.jsx(p,{path:"/security",element:N(["cto","founder","admin"],n.jsx(Mh,{}))}),n.jsx(p,{path:"/optimizer",element:N(["cto","founder","admin"],n.jsx(qh,{}))}),n.jsx(p,{path:"/executive-insights",element:N(["investor","founder","admin"],n.jsx(Dh,{}))}),n.jsx(p,{path:"/global-intelligence",element:N(["founder","admin","investor"],n.jsx(Qh,{}))}),n.jsx(p,{path:"/partner-intelligence",element:N(["founder","admin","partner","sales"],n.jsx(Vh,{}))}),n.jsx(p,{path:"/finops-intelligence",element:N(["founder","admin","finance"],n.jsx(Hh,{}))}),n.jsx(p,{path:"/compliance-center",element:N(["founder","admin"],n.jsx(Wh,{}))}),n.jsx(p,{path:"/executive-command",element:N(["founder","admin"],n.jsx(Kh,{}))}),n.jsx(p,{path:"/quality-gate",element:N(["founder","admin"],n.jsx($h,{}))}),n.jsx(p,{path:"/knowledge-graph",element:N(["investor","founder","admin"],n.jsx(Yh,{}))}),n.jsx(p,{path:"/partner-portal",element:N(["partner","founder","admin"],n.jsx(Zh,{}))}),n.jsx(p,{path:"/customer-portal",element:N(["enterprise_client","founder","admin"],n.jsx(Jh,{}))}),n.jsx(p,{path:"/verification",element:N(["founder","admin"],n.jsx(eg,{}))}),n.jsx(p,{path:"/audit-log",element:N(["founder","admin"],n.jsx(rg,{}))}),n.jsx(p,{path:"/ai-governance",element:N(["founder","admin","cto"],n.jsx(ig,{}))}),n.jsx(p,{path:"/ai-performance",element:N(["founder","admin","cto","finance","investor"],n.jsx(sg,{}))}),n.jsx(p,{path:"/firecrawl-health",element:N(["founder","admin"],n.jsx(mg,{}))}),n.jsx(p,{path:"/referrals",element:n.jsx(tg,{})}),n.jsx(p,{path:"/email-monitor",element:N(["marketing","founder","admin"],n.jsx(ng,{}))}),n.jsx(p,{path:"/deliverability",element:N(["marketing","founder","admin"],n.jsx(pg,{}))}),n.jsx(p,{path:"/campaign-preview",element:N(["marketing","founder","admin"],n.jsx(fg,{}))}),n.jsx(p,{path:"/prospect-targets",element:N(["marketing","founder","admin"],n.jsx(ag,{}))}),n.jsx(p,{path:"/outreach-dashboard",element:N(["marketing","founder","admin"],n.jsx(og,{}))}),n.jsx(p,{path:"/user-analytics",element:N(["sales","marketing","founder","admin"],n.jsx(Lh,{}))}),n.jsx(p,{path:"/war-room-yc",element:N(["founder","admin"],n.jsx(Fh,{}))}),n.jsx(p,{path:"/ai-workforce",element:n.jsx(Nh,{})}),n.jsx(p,{path:"/ai-workforce/onboarding/:id",element:n.jsx(_h,{})}),n.jsx(p,{path:"/ai-workforce/chat/:id",element:n.jsx(zh,{})}),n.jsx(p,{path:"/workforce-marketplace",element:n.jsx(bo,{})}),n.jsx(p,{path:"/hire",element:n.jsx(bo,{})}),n.jsx(p,{path:"/workforce-dashboard",element:n.jsx(jh,{})}),n.jsx(p,{path:"/workforce-connectors",element:n.jsx(Gh,{})}),n.jsx(p,{path:"/workforce-router",element:n.jsx(Bh,{})}),n.jsx(p,{path:"/workforce-safety",element:n.jsx(Uh,{})}),n.jsx(p,{path:"*",element:n.jsx(Op,{})})]}),n.jsx(p,{path:"/widget/iao",element:n.jsx(Wf,{})}),n.jsx(p,{path:"/themis/verify",element:n.jsx(ho,{})}),n.jsx(p,{path:"/themis/verify/:hash",element:n.jsx(ho,{})}),n.jsx(p,{path:"/verify/benchmark",element:n.jsx(un,{})}),n.jsx(p,{path:"/verify/benchmark/:hash",element:n.jsx(un,{})}),n.jsx(p,{path:"/benchmark/verify",element:n.jsx(un,{})}),n.jsx(p,{path:"/benchmark/verify/:hash",element:n.jsx(un,{})})]})}),!a&&e&&n.jsx(i.Suspense,{fallback:null,children:n.jsx(hp,{})}),!a&&n.jsxs(n.Fragment,{children:[n.jsx(i.Suspense,{fallback:null,children:n.jsx(gp,{})}),n.jsx(i.Suspense,{fallback:null,children:n.jsx(bp,{})}),n.jsx(i.Suspense,{fallback:null,children:n.jsx(yp,{})}),n.jsx(i.Suspense,{fallback:null,children:n.jsx(vp,{})})]})]})})]})})})})})};try{const e=document.getElementById("root");No(e).render(n.jsx(yg,{})),requestAnimationFrame(()=>{requestAnimationFrame(()=>{const t=document.getElementById("app-loader");t&&t.remove();const a=document.createElement("div");a.id="react-mounted",a.style.display="none",document.body.appendChild(a);const o=document.createElement("link");o.rel="stylesheet",o.href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700;
1019800&family=JetBrains+Mono:wght@400;500&display=swap",document.head.appendChild(o)})})}catch(e){console.error("Fatal mount error:",e);const t=document.getElementById("app-loader");t&&(t.innerHTML='<div style="color:#0F172A;font-family:Inter,system-ui,sans-serif;text-align:center;padding:20px;max-width:480px;"><p style="font-size:20px;font-weight:800;margin-bottom:8px;letter-spacing:0.15em;text-transform:uppercase;"><span style="color:#4F46E5;">INFERNEX</span> AI</p><p style="font-size:12px;color:#475569;word-break:break-all;margin-bottom:16px;">'+(e instanceof Error?e.message:String(e))+'</p><button onclick="location.reload(true)" style="padding:10px 28px;background:#4F46E5;color:#ffffff;border:none;border-radius:8px;cursor:pointer;font-size:14px;font-weight:700;">Refresh</button></div>')}export{$l as $,cs as A,us as B,eb as C,Km as D,hs as E,Mg as F,Xg as G,Hg as H,qm as I,Wg as J,Kg as K,rr as L,gs as M,$g as N,ms as O,Ym as P,fs as Q,op as R,Pm as S,ar as T,ys as U,tp as V,Ug as W,xn as X,wn as Y,vs as Z,f as _,te as a,Bl as a0,Rg as a1,Xn as a2,Ce as a3,Sn as a4,jl as a5,Gl as a6,Ir as a7,ec as a8,Ql as a9,me as aa,Og as ab,ps as ac,Lm as ad,Kt as ae,oc as af,Qr as ag,Ur as ah,ap as ai,um as aj,Nc as ak,qg as al,Ju as am,Qg as an,Nr as ao,ds as ap,zg as aq,Nl as ar,Tg as as,Eg as at,_l as au,Vg as av,Bg as aw,_g as ax,Dg as ay,tb as az,zm as b,pe as c,zr as d,jr as e,Gr as f,Br as g,ia as h,ma as i,Ht as j,Lt as k,Wn as l,Vr as m,Jg as n,Zg as o,ir as p,Lg as q,Ng as r,ye as s,Vm as t,Pg as u,Fg as v,Yg as w,Gg as x,se as y,jg as z};
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.