54 versions

• feat(usage): record a status code on usage logs so failures can be classified (#470)latest○786e53532026-08-03 11:26
• feat(search): add a billed POST /v1/search pass-through (#473)○20e508d92026-08-03 11:16
• feat(usage): break spend down by session, endpoint, and provider (#469)○169d1b3e62026-08-03 11:03
• feat(aliases): scope a model alias to a single user (#455)○4dbbfebc2026-07-31 00:54
• fix(dashboard): distinguish a missing /models endpoint from an unreachable provider (#454)○5ab4d3ca2026-07-31 00:41
• feat(auth): rate-limit dashboard sign-in against brute-force attempts (#437)no API changes2657bc92026-07-28 07:28
• feat(dashboard): shared table framework with filtering, pagination, selection, and bulk actions (#415)○10c8e09b32026-07-24 12:56
• feat(usage): ingest external usage events for API-equivalent cost tracking (#392)○23bf90e942026-07-23 22:59
• feat(dashboard): disable adding providers when OTARI_SECRET_KEY is unset (#391)○2661aafc2026-07-23 19:49
• fix(dashboard): persist sign-in with an HttpOnly session cookie (#384)○2f462b032026-07-23 16:48
• fix(dashboard): make the provider picker open instantly (lazy split) (#382)▲1●3○1e16e49f2026-07-23 16:42
• feat(files): stream uploads and downloads instead of buffering in memory (#380)no API changescbee20e2026-07-23 13:14
• feat(pricing): meter cache writes and long context (#379)○1875592572026-07-23 10:56
• feat(pricing): add reviewable price refresh (#367)○3e5bdbb82026-07-22 19:22
• feat: add dashboard key lifecycle controls (#360)○43f31c9c2026-07-22 12:38
• feat: price cached input tokens in standalone cost calculation (#356)○141a8196e2026-07-22 12:14
• feat(dashboard): built-in tools & guardrails configuration (#306) (#350)○335991e22026-07-22 00:20
• feat(dashboard): expand Settings with a full config viewer, wider settable set, and search (#346)○15d6d8ccf2026-07-21 20:58
• fix(providers): make API key optional when the provider env var is set (#344)○1f7b96c52026-07-21 20:33
• feat(dashboard): usage & analytics page with aggregation endpoints (#303) (#345)○22c95f652026-07-21 20:17
• feat(dashboard): provider health monitor on the Providers page (#347)○1769c7132026-07-21 20:09
• feat(dashboard): activity / request log viewer with per-request latency (#330)○6631a0062026-07-21 10:33
• feat(dashboard): budget and user management with two-layer model access (#322)○25eac679e2026-07-21 10:12
• feat(dashboard): API key management and per-key model access control (#318)○12f49d70f2026-07-20 14:26
• feat(auth): accept x-api-key header so Anthropic-native clients work (fixes #253) (#315)○614b4bd962026-07-20 10:13
• feat(ui): manage providers, pricing, and aliases at runtime from the dashboard (#297)○77d092762026-07-20 09:28
• feat(ui): add a standalone admin dashboard for models, pricing, and settings (#227)○12fc023902026-07-17 01:20
• fix(deps): bump any-llm to 1.21.0 to meter streaming /v1/messages usage (#292)○101335b22026-07-16 17:16
• fix(api): unify error handling and hybrid streaming fallback across completion routes (#281)no API changesff6acb12026-07-16 14:05
• fix(batches): enforce user resolution, budget, rate limiting, and batch ownership (#273)○13f2c23e2026-07-16 13:51
• feat: add API key rotation endpoint (#285)○12756f082026-07-16 13:05
• fix(openapi): add ApiKeyAuth security scheme to schema definition (#248)○45ca5608c2026-07-14 15:15
• perf: async MCP/guardrail URL checks, dedupe provider resolution (#250)no API changes878d12a2026-07-14 10:54
• feat(hybrid): forward request session_label to platform usage report (#233)○35fdbe752026-06-30 16:21
• feat(usage): forward provider cached-token counts in usage reports (#196)○21b6a80e2026-06-23 13:07
• refactor: name the runtime mode "hybrid" instead of "platform" (#163)no API changes3d4c9b12026-06-23 10:53
• feat(api): derive request schemas from any-llm Params to stop param drift (#160)▲1○36d3948a12026-06-15 14:22
• Rename "Otari Gateway" to "Otari" (#142)no API changes8cd4f7f2026-06-12 18:20
• feat(files): file uploads + document understanding for local models○556ba31e2026-06-10 08:55
• fix(messages): address review on count_tokens endpoint▲1○3691ac522026-06-09 13:49
• feat(messages): add /v1/messages/count_tokens for Claude Code support○12664e112026-06-09 13:27
• feat(guardrails): default to PIGuard via encoderfile, monitor mode○316632cd2026-06-03 09:30
• fix(guardrails): address Copilot reviewno API changesd1346632026-06-02 17:02
• chore(openapi): regenerate spec for guardrails field○3d3532252026-06-02 15:58
• fix(ci): typecheck async-generator aclose + regenerate openapi spec○469fc2f92026-06-01 11:05
• fix(messages,responses): streaming provider errors → HTTP error + regen OpenAPI spec○91e3273f2026-05-28 11:16
• fix: address Copilot review feedback on model discovery○27af67732026-05-21 09:55
• feat: auto-discover models from configured providers for /v1/models▲2○80c95f3b2026-05-21 07:56
• Merge pull request #64 from mozilla-ai/feat/mcp-server-id-resolve○1c9dc4d22026-05-21 13:57
• Regenerate OpenAPI spec with /v1/rerank endpoint●1○7d624ccb2026-04-17 20:11
• chore(openapi): regenerate spec for mcp_server_ids field▲6○6ca91a6b2026-05-14 16:22
• fix: address review feedback on PR #52▲1○3ce5d7c92026-05-20 14:33
• Regenerate OpenAPI spec for /v1/moderations route▲28●60○597a799d62026-04-17 14:01
• 0.0.0-dev2026-07-17 12:07