All documentation
Browse the published guides, architecture explanations and API reference.
ArchitectureArchitecture and request lifecycledocs/design/README.mdArchitectureRouting and adaptive selection designdocs/design/adaptive-routing.mdArchitectureAuthentication and identity designdocs/design/authentication.mdArchitecturePolicy-based authorization: scopes, grants and inheritancedocs/design/authorization.mdArchitectureExact and semantic cache designdocs/design/cache.mdArchitectureGuardrail designdocs/design/guardrails.mdArchitecturePolicy service designdocs/design/policy-service.mdArchitectureRate limiting, budgets and settlement designdocs/design/rate-limiting.mdDocumentationLimits, budgets, and the other controlsdocs/gateway-controls.mdDocumentationGuardrailsdocs/guardrails.mdDocumentationHow-to guidesdocs/how-to/README.mdConnect and configureAdd a provider and a modeldocs/how-to/add-a-provider-and-a-model.mdConnect and configureAdd a second key to a provider, and rotate onedocs/how-to/add-a-second-provider-key.mdAccess and controlsBuild a team hierarchydocs/how-to/build-a-team-hierarchy.mdConnect and configureConfigure Pointsman without guessingdocs/how-to/configure-pointsman.mdOperatePrepare a deploymentdocs/how-to/deploy.mdOperateWork out what happened to a requestdocs/how-to/diagnose-a-request.mdAccess and controlsIssue a key and give it a budgetdocs/how-to/issue-keys-and-set-budgets.mdLearning · experimentalJev task routing (experimental, opt-in)docs/how-to/jev-routing.mdLearning · experimentalRoute only to models with quality evidence (experimental, opt-in)docs/how-to/quality-guarded-routing.mdConnect and configureSend some of your traffic to a different modeldocs/how-to/split-traffic-between-models.mdAccess and controlsTune the semantic cachedocs/how-to/tune-the-semantic-cache.mdLearning · experimentalTurn on outcome learningdocs/how-to/turn-on-outcome-learning.mdAccess and controlsWrite and enforce a guardraildocs/how-to/write-and-enforce-a-guardrail.mdOperateKnown limitationsdocs/limitations.mdOperateOperating notesdocs/operating-notes.mdDocumentationJournal epoch directoriesdocs/operations/journal-epochs.mdDocumentationLearning history: window, maintenance and recovery (learner)docs/operations/learning-history.mdRelease scopeLatency and benchmark breakdowndocs/performance.mdGet startedPointsman documentationdocs/public-index.mdDocumentationReferencedocs/reference/README.mdAPI referenceAdmin API referencedocs/reference/admin-api.mdAPI referenceApplication APIdocs/reference/client-api.mdAPI referenceChat Completions compatibilitydocs/reference/compatibility.mdAPI referenceConfiguration referencedocs/reference/configuration.mdDocumentationDashboard configuration ownershipdocs/reference/dashboard-configuration.mdAPI referenceOperational limits and tuningdocs/reference/operational-limits.mdAPI referenceResponse headersdocs/reference/response-headers.mdRelease scopeRelease scope and qualificationdocs/release-status.mdConnect and configureRouting strategiesdocs/routing-strategies.mdOperateSecurity posturedocs/security.mdDocumentationThe semantic cachedocs/semantic-cache.mdGet startedQuickstart: your first Chat Completiondocs/tutorial/README.mdGet startedA guided tour of the dashboarddocs/tutorial/dashboard.mdGet startedTry Pointsman with mock providersdocs/tutorial/mock-provider.md