fix(a2a)!: bill completion-bridge agents at their own prices on /a2a send and stream ( #49150 )
2026-10-30 19:43:45 -07:00
feat(accounting): add accounting contracts, PricingSnapshot and ledger schema (LIT-10193) ( #48862 )
2026-10-28 14:39:25 -07:00
fix(anthropic_endpoints)!: return Anthropic error envelopes on /v1/messages auth failures ( #48466 )
2026-11-03 01:53:29 +00:00
feat(accounting): capture request usage through the durable meter ( #49133 )
2026-10-31 07:32:25 -07:00
feat(hosted_vllm)!: keep vLLM's stop_reason and token_ids on chat choices and stream chunks ( #48393 )
2026-11-02 13:26:29 -08:00
fix(logging): keep the caller's call_type when chat completions are bridged to the Responses API ( #49710 )
2026-11-03 10:02:54 +00:00
refactor(typing): clear 1181 basedpyright Any errors across 431 files ( #47789 )
2026-10-24 11:31:09 -07:00
feat(realtime)!: support latest OpenAI audio models ( #35600 )
2026-11-03 07:27:52 +00:00
fix(vector-stores): keep proxy context out of provider vector store metadata ( #47021 )
2026-10-17 12:51:47 -07:00
fix(decisions)!: honor a deployment's own supports_audio_input and supports_video_input flags ( #49207 )
2026-10-31 01:09:46 -07:00
feat(vertex-ai)!: support gemini 3.1 flash tts ( #31915 )
2026-10-21 04:18:42 +00:00
fix(managed_files): keep the upload the managed id encodes when two target models share a deployment ( #49420 )
2026-11-03 00:58:41 +00:00
fix(mcp): load deprecated package aliases on demand ( #48615 )
2026-10-27 19:46:19 +00:00
feat(extraction): add /v1/extraction with Cloudflare Clairic support ( #47473 )
2026-10-20 15:51:45 -07:00
fix(hosted_vllm): accept the vLLM production-stack router's file and batch objects ( #49068 )
2026-10-30 23:11:34 -07:00
fix(fine_tuning): stop forwarding api_version to the OpenAI client ( #44995 )
2026-10-21 01:58:25 +00:00
fix(proxy): keep the litellm command light so spawned workers answer uvicorn's first health ping ( #45039 )
2026-10-29 10:25:14 -07:00
fix(google_genai): read one-object and null spellings of repeated fields in countTokens and generateContent ( #48771 )
2026-10-28 21:46:06 -07:00
fix(streaming)!: send the include_usage chunk with an empty choices array ( #47109 )
2026-10-21 02:26:05 +00:00
feat(dashscope): add image edits and text-to-speech ( #48998 )
2026-10-29 17:48:38 -07:00
test(mcp): cancel disconnected HTTP tool handlers ( #49436 )
2026-10-31 16:41:49 -07:00
fix(anthropic)!: carry prompt_cache_breakpoint to cache_control on chat completions to Claude ( #49783 )
2026-11-03 12:56:22 +00:00
fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF ( #48755 )
2026-10-28 17:00:29 -07:00
fix(logging): keep the caller's call_type when chat completions are bridged to the Responses API ( #49710 )
2026-11-03 10:02:54 +00:00
feat(proxy): replace the pre-migration row check with a rolled-back dry run of the pending migrations ( #48961 )
2026-10-29 09:39:26 -07:00
fix(e2e): restore main qualification across providers and UI ( #49763 )
2026-11-03 20:38:47 +00:00
refactor(mcp): drop the experimental label from the MCP server and client paths ( #48513 )
2026-10-27 17:47:50 +00:00
fix(proxy): allow model instances in LiteLLM_EndUserTable validation ( #41677 )
2026-10-15 10:57:26 -07:00
test(rust-bridge): relocate dispatch tests before routing changes ( #46926 )
2026-10-16 12:29:33 -07:00
fix(vllm): send the pass-through key as a bearer on both /vllm routes ( #46861 )
2026-10-30 08:57:48 -07:00
fix(proxy): keep the Codex catalog key lookup in line with the call path and cap each catalog fetch
2026-11-04 22:05:34 -08:00
fix(rag)!: use team provider credentials and allow viewer and JWT team access ( #45497 )
2026-10-17 23:51:48 +00:00
feat(realtime)!: support latest OpenAI audio models ( #35600 )
2026-11-03 07:27:52 +00:00
refactor(proxy): inject gateway activity through a typed repository ( #49116 )
2026-10-30 16:25:11 -07:00
fix(cache_control)!: keep one deployment's prompt-cache opt-in from reaching siblings on the same model ( #49685 )
2026-11-03 07:55:51 +00:00
feat(decisions): add Parallel AI web-grounded decisions ( #49188 )
2026-11-03 04:47:02 +00:00
fix(anthropic)!: carry prompt_cache_breakpoint to cache_control on chat completions to Claude ( #49783 )
2026-11-03 12:56:22 +00:00
fix(messages): make rust messages copy python messages behaviour ( #49611 )
2026-11-02 21:02:52 -08:00
fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF ( #48755 )
2026-10-28 17:00:29 -07:00
fix(search): price Tavily advanced and Linkup deep searches at their own cost map entries ( #47411 )
2026-10-19 23:32:31 -07:00
fix(proxy): let a config license unlock enterprise settings and spell "environment" right in license errors ( #49745 )
2026-11-03 09:56:08 +00:00
feat(skills): add a litellm-setup agent skill ( #47731 )
2026-10-28 15:44:43 -07:00
fix(databricks): reuse base OAuth for decisions ( #48395 ) ( #48527 )
2026-10-26 20:18:16 -07:00
fix(bedrock)!: fail a Nova Sonic realtime session whose Bedrock stream drops instead of closing 1000 ( #49170 )
2026-11-03 07:56:23 +00:00
feat(hosted_vllm)!: keep vLLM's stop_reason and token_ids on chat choices and stream chunks ( #48393 )
2026-11-02 13:26:29 -08:00
fix(proxy): cooldown errors carry 429, Sentry starts at boot without request bodies, tracing failures log at ERROR, exporter counters exported ( #46344 )
2026-10-16 00:53:52 +00:00
fix(responses)!: authorize and decode comparison response IDs ( #49402 )
2026-11-03 03:40:46 +00:00
fix(exceptions)!: honor status code when mapping OpenAI invalid_request_error exceptions ( #32537 )
2026-10-29 08:05:47 -07:00
feat(videos): serve custom_provider_map BaseVideoConfig providers on /v1/videos ( #49785 )
2026-11-02 21:28:46 -08:00
feat(vector_stores): add OpenSearch k-NN search as a vector store provider ( #47759 )
2026-10-29 08:15:05 -07:00
fix(proxy): answer Codex inside its catalog deadline and key the catalog cache without hashing the OpenAI key
2026-11-04 14:44:18 -08:00
fix(anthropic)!: drop beta-gated body fields under drop_params ( #48793 )
2026-10-29 06:25:31 -07:00
feat(anthropic): pick the beta allowlist row per deployment and surface input_transformations ( #49689 )
2026-11-03 01:55:26 +00:00
docs(agents): keep directory rules in nested AGENTS.md and pin integration tests to CircleCI ( #46796 )
2026-10-16 14:57:49 -07:00
ci: run the integration groups on GitHub Actions ( #48985 )
2026-10-29 12:50:08 -07:00
fix(bedrock): omit system prompt from audio transcription and report its token usage ( #41565 )
2026-10-28 19:23:15 -07:00
feat(models): add Cerebras Llama 3.1 hyphenated aliases (cerebras/llama-3.1-8b, cerebras/llama-3.1-70b) ( #48157 ) ( #48158 )
2026-10-24 17:16:33 -07:00
refactor(mcp): drop the experimental label from the MCP server and client paths ( #48513 )
2026-10-27 17:47:50 +00:00
feat(realtime)!: support latest OpenAI audio models ( #35600 )
2026-11-03 07:27:52 +00:00
test(integration): hand every harness server a port from one per-worker block ( #46761 )
2026-10-15 16:56:40 -07:00
ci(circleci): clone the public repo over HTTPS so main jobs stop failing at checkout ( #49394 )
2026-10-31 02:53:54 -07:00
refactor(mcp): drop the experimental label from the MCP server and client paths ( #48513 )
2026-10-27 17:47:50 +00:00
fix(router)!: bound fake-streamed calls by the request timeout, not stream_timeout ( #47162 )
2026-10-17 22:08:02 +00:00
refactor(router): extract PythonCore behind RouterCore and pass the router its config ( #49119 )
2026-10-31 12:54:23 -07:00
fix(router): cap the default OTPM reservation and reconcile token counters after each call ( #49054 )
2026-10-29 20:32:04 -07:00
feat(realtime)!: support latest OpenAI audio models ( #35600 )
2026-11-03 07:27:52 +00:00
fix(cost_map): mark fireworks_ai/minimax-m3 as vision and pin capabilities verified against live calls ( #43390 )
2026-10-29 00:23:11 -07:00
fix(responses)!: count input tokens the way inference sends the request ( #47363 )
2026-10-20 15:13:59 +00:00
feat(dashscope): add image edits and text-to-speech ( #48998 )
2026-10-29 17:48:38 -07:00
feat(lite): offer each model group's reasoning efforts in pi, OpenCode and Codex ( #48949 )
2026-10-29 18:15:44 -07:00
fix(cost-map): rolling registry audit, add elevenlabs api rates, together minimax-m3.1-flash, mistral voxtral 3 and azure_ai phi-4 dates ( #47623 )
2026-10-23 14:57:35 -07:00
fix: honor empty mock completion responses ( #48162 ) ( #48266 )
2026-10-28 00:56:45 +00:00
fix(openai): keep the upstream error body on e.response for OpenAI and Azure errors ( #49079 )
2026-10-29 23:30:02 -07:00
feat(gemini)!: auto-continue gemini-4-argon long decodes with continuationToken ( #46240 )
2026-10-15 22:54:23 +00:00
feat(proxy): add LITELLM_OFFLINE_MODE, one switch for the fetches LiteLLM makes on its own ( #48096 )
2026-10-23 16:54:19 -07:00
chore(hooks): drop Conventional Branches pre-push hook in favor of litellm_ prefix ( #44951 )
2026-10-20 00:35:39 +00:00
fix(streaming): keep merged think tags balanced and split SSE lines only on CR/LF ( #48755 )
2026-10-28 17:00:29 -07:00
Add guardrails.apply SDK helper ( #49217 ) ( #49469 )
2026-10-31 23:37:45 +00:00
feat(proxy)!: model family and access group entries in model_max_budget ( #47722 )
2026-10-28 15:40:57 -07:00
fix(health)!: openai realtime health check reports healthy with a missing or invalid api key ( #41233 )
2026-10-29 00:40:02 -07:00
fix(ci): measure PR gates from the target branch's current tip ( #49078 )
2026-10-29 18:28:16 -07:00
fix(proxy): keep the litellm command light so spawned workers answer uvicorn's first health ping ( #45039 )
2026-10-29 10:25:14 -07:00
feat(proxy)!: add suppress_startup_output to turn off startup prints ( #47708 )
2026-10-29 03:47:04 -07:00
feat(realtime)!: support latest OpenAI audio models ( #35600 )
2026-11-03 07:27:52 +00:00
fix(cost-map): price Mistral Large 4 at its list price now that the launch sale ended ( #48825 )
2026-10-28 18:12:55 +00:00
refactor(router): move router callers onto the entry points and invert the proxy imports ( #48847 )
2026-10-29 17:26:33 +00:00
feat(realtime)!: support latest OpenAI audio models ( #35600 )
2026-11-03 07:27:52 +00:00
refactor(router): route the router family through RoutingContext and close the outside surface ( #49118 )
2026-10-30 19:18:54 -07:00
fix(ci): build base native extension for OpenAPI lint ( #49136 )
2026-10-29 16:56:44 -07:00
chore(deps): require anyio>=4.14.0 to rule out the lock-waiter deadlock ( #47378 )
2026-10-19 12:37:47 -07:00
ci: run the dashboard API types and Terraform endpoint checks on every change that can alter the spec ( #49022 )
2026-10-29 07:19:06 -07:00
refactor(router): route the router family through RoutingContext and close the outside surface ( #49118 )
2026-10-30 19:18:54 -07:00
fix(register_model)!: keep bedrock geo deployments priced across price data reloads ( #41523 )
2026-10-17 19:10:04 +00:00
feat(skills): add a litellm-setup agent skill ( #47731 )
2026-10-28 15:44:43 -07:00
fix(responses)!: authorize and decode comparison response IDs ( #49402 )
2026-11-03 03:40:46 +00:00
fix(responses): bill a background response once, on the first read or poll that sees it finished ( #45703 )
2026-10-15 11:00:14 -07:00
fix(streaming)!: map stream read timeouts to litellm.Timeout on chat, messages, and responses ( #47198 )
2026-10-18 13:58:34 +00:00
feat(model_info): add supports_continuation capability flag for Gemini long decode continuation ( #49500 )
2026-11-02 13:53:58 -08:00
fix(router): keep the prompt caching affinity pin alive as long as the prefix's cache_control ttl ( #46084 )
2026-10-14 21:10:09 -07:00
fix(images): honor max_retries and prevent Router retry amplification ( #48705 )
2026-10-28 10:46:47 -07:00
fix(proxy): unset DB environment variables removed from the row on every worker by row revision ( #46709 )
2026-10-17 03:51:05 +00:00
refactor(router): extract PythonCore behind RouterCore and pass the router its config ( #49119 )
2026-10-31 12:54:23 -07:00
refactor(router): route the router family through RoutingContext and close the outside surface ( #49118 )
2026-10-30 19:18:54 -07:00
refactor(router): extract PythonCore behind RouterCore and pass the router its config ( #49119 )
2026-10-31 12:54:23 -07:00
fix(scheduler): read shared queue from Redis so expired dead-replica entries stop blocking waiters ( #49387 )
2026-10-31 11:43:46 -07:00
refactor(rust-bridge): require the native extension ( #48852 )
2026-10-28 13:16:56 -07:00
fix(ci): measure PR gates from the target branch's current tip ( #49078 )
2026-10-29 18:28:16 -07:00
fix(ci): flag any schema.prisma change inside model LiteLLM_SpendLogs in the spend logs guard (LIT-10240) ( #48905 )
2026-10-28 15:52:15 -07:00
fix(anthropic)!: send /v1/messages PDF documents to OpenAI-shaped providers as file parts ( #47166 )
2026-10-29 08:06:12 -07:00
fix(prompt_templates): do not mutate messages in-place in function_call_prompt ( #41676 )
2026-10-15 14:37:11 -07:00
refactor(cost-map): drop the legacy max_tokens field in favor of max_input_tokens and max_output_tokens ( #48471 )
2026-10-26 17:03:05 -07:00
fix(responses): replay the original Responses item order, phase, and boundaries on the next chat turn ( #48912 )
2026-10-29 00:56:02 -07:00
fix(cost): price typesafe/jev-1.14.0 ( #46594 )
2026-10-14 20:27:35 -07:00
feat(model-catalog): read successor_model, the provider's recommended replacement for a deprecating row ( #49791 )
2026-11-03 11:51:44 -08:00
fix(bedrock): price claude_platform calls from their own cost map rows ( #48150 )
2026-10-24 13:54:18 -07:00
fix(openai): warn when the GPT-5 reasoning effort gate drops a level under drop_params ( #47761 )
2026-10-29 04:46:29 -07:00
fix: support Python 3.15 ( #48359 )
2026-10-26 16:00:11 +00:00
refactor(router): move router callers onto the entry points and invert the proxy imports ( #48847 )
2026-10-29 17:26:33 +00:00
fix(router)!: stop a chat-mode Mantle deployment from rewriting the shared Responses routing row ( #46263 )
2026-10-17 21:39:14 +00:00
