What's Changed
- feat: calculate
total_tokensmanually if it is missing and can be calculated from input and output tokens by @wileykestner in #18445 - feat: Add built-in migration lock to prevent concurrent Prisma migrate deploy by @minijeong-log in #14440
- feat: Add Prometheus metrics for request queue time and guardrails by @hamzaq453 in #17973
- Revert "feat: Add built-in migration lock to prevent concurrent Prisma migrate deploy" by @krrishdholakia in #18719
- Add Anthropic cache control option to image tool call results by @Mr-Pepe in #18674
- Hotfix - docs qualifire by @drorIvry in #18724
- fix(proxy): return json error response instead of sse format for initial streaming errors by @krisxia0506 in #18757
- feat(prometheus): add caching metrics for cache hits, misses, and tokens by @Harshit28j in #18755
- Litellm embeddings calltype fix for guardrail precallhook by @kothamah in #18740
- fix: prevent duplicate User-Agent tags in request_tags by @Tianduo16 in #18723
- make base_connection_pool_limit default value the same by @wilsonjr in #18721
- feat: Add abliteration.ai provider by @abliteration-ai in #18678
- fix(braintrust): pass span_attributes in async logging and skip tags on non-root spans by @ihower in #18409
- OpenRouter embeddings API support by @elkkhan in #18391
- fix(gemini): support snake_case for google_search tool parameters by @Chesars in #18451
- fix(proxy): use async anthropic client to prevent event loop blocking by @runixer in #18435
- fix: properly use litellm api keys by @lucasrothman in #18832
- fix: add idx on LOWER(user_email) for faster duplicate email checks l… by @Harshit28j in #18828
- feat(tag-routing): support toggling tag matching between ANY and ALL by @choby-shun in #18776
- fix: Mask extra header secrets in model info by @Chesars in #18822
- fix: add xiaomi_mimo to LlmProviders enum to fix router support by @Chesars in #18819
- docs: fix PDF documentation inconsistency in Anthropic page by @Chesars in #18816
- fix(workflow): Update issue labeling with working regex pattern by @Chesars in #18821
- fix: proactive RDS IAM token refresh to prevent 15-min connection failed by @Harshit28j in #18795
- feat(ui): add custom proxy base URL support to Playground by @Chesars in #18661
- [Fix] Normalize Proxy Config Callback by @yuneng-jiang in #18775
- [Feature] Key and Team Router Setting by @yuneng-jiang in #18790
- [Infra] Fixing UI Build by @yuneng-jiang in #18835
- [feat] Focus export support by @uc4w6c in #18802
- [docs] add focus by @uc4w6c in #18837
- feat: added qualifire eval webhook by @drorIvry in #18836
- [fix] how to execute cloudzero sql by @uc4w6c in #18841
- [FEAT]: Add support for Vertex AI API keys by @Sameerlite in #18806
- [feat] add mcp registry by @uc4w6c in #18850
- fix: Improve error messages and validation for wildcard routing with multiple credentials by @akraines in #18629
- Fix: Add thought_signatures to VertexGeminiConfig and test by @jutaz in #18853
- Staging - 01/07/2026 by @krrishdholakia in #18763
- Litellm staging 01 08 2026 by @krrishdholakia in #18833
- Staging 01/06/2026 by @krrishdholakia in #18715
- feat: add Bedrock as a backend API for token counting by @raghav-stripe in #18858
- [Fix] Nova model detection for Bedrock provider (#17910) by @sjmatta in #18250
- fix(security): prevent expired key plaintext leak in error response by @Harshit28j in #18860
- [Infra] UI - Unit Test: Adding Tests to Expand Coverage by @yuneng-jiang in #18848
- [Infra] UI - E2E Test: Refactor Page Settings + Test for Page Navigation by @yuneng-jiang in #18849
- fix: prevent Prisma migration workflow from running in forks by @Chesars in #18863
- fix(azure): add logprobs support for Azure OpenAI GPT-5.2 model by @andres-ortizl in #18856
- Fix: response_format leaking into extra_body by @Sameerlite in #18859
- Fix/litellm sdk embedding headers missing field by @gauthiermartin in #18844
- perf: 92.7% faster provider config lookup - LiteLLM now stresses LLM providers 2.5× more by @AlexsanderHamir in #18867
- Fix: google_genai streaming adapter provider handling by @mel2oo in #18845
- docs: Update header to be markdown bold by removing space by @DominikHallab in #18846
- fix: align max_tokens with max_output_tokens for consistency by @Chesars in #18820
- [Fix] Watsonx Audio Transcription - filter model field by @shivamrawat1 in #18810
- [Fix] Enforce org level max budget by @shivamrawat1 in #18813
- feat: update prices json for novita provider by @ji-chuan-che in #18540
- Add fix for bedrock_cache, metadata and max_model_budget by @Harshit28j in #18872
- Fix :test_count_tokens_caching by @Sameerlite in #18871
- [Fix] UI Login Case Sensitivity by @yuneng-jiang in #18877
- [fix] mcp error in multiple servers by @uc4w6c in #18855
- [Infra] Router Fields Endpoint + React Query for Router Fields by @yuneng-jiang in #18880
- Add memory leak detection tests with CI integration by @AlexsanderHamir in #18881
- [Bug]: Add Custom CA certificates to boto3 clients by @Sameerlite in #18852
- [Infra] UI - E2E Test: New DB Branch Per Test Run by @yuneng-jiang in #18885
- [Feature] UI - Endpoint Activity in Usage by @yuneng-jiang in #18798
- [Infra] Building UI for QA Testing by @yuneng-jiang in #18894
- [Fix] turn_off_message_logging Does Not Redact Request Messages in proxy_server_request Field When Stored to Database by @ishaan-jaff in #18897
- update OpenRouter docs to include embedding support by @elkkhan in #18874
- [Infra] Proxy Extras CI/CD Fix by @yuneng-jiang in #18901
- [Feat] Manus API - Add GET Responses API Support by @ishaan-jaff in #18900
- [Docs] v1.80.13.rc.1 Docs by @yuneng-jiang in #18906
- [Feat] Manus FILES API - Add File upload, get, delete, list by @ishaan-jaff in #18904
- Revert "fix: Improve error messages and validation for wildcard routing with multiple credentials" by @ishaan-jaff in #18907
- fix: #18629 by @akraines in #18909
- [Fix] Litellm fix org buget enforcement tests by @shivamrawat1 in #18908
- [Fix] UI - Endpoint Activity Trend X-Axis and Time Range by @yuneng-jiang in #18911
- [Docs] Endpoint Usage Docs by @yuneng-jiang in #18913
- [Infra] Building UI by @yuneng-jiang in #18914
- perf release notes by @AlexsanderHamir in #18915
- Add ScrapeGraph MCP server configuration by @VinciGit00 in #18923
New Contributors
- @wileykestner made their first contribution in #18445
- @minijeong-log made their first contribution in #14440
- @Mr-Pepe made their first contribution in #18674
- @Tianduo16 made their first contribution in #18723
- @wilsonjr made their first contribution in #18721
- @abliteration-ai made their first contribution in #18678
- @ihower made their first contribution in #18409
- @elkkhan made their first contribution in #18391
- @runixer made their first contribution in #18435
- @choby-shun made their first contribution in #18776
- @jutaz made their first contribution in #18853
- @sjmatta made their first contribution in #18250
- @andres-ortizl made their first contribution in #18856
- @gauthiermartin made their first contribution in #18844
- @mel2oo made their first contribution in #18845
- @DominikHallab made their first contribution in #18846
- @ji-chuan-che made their first contribution in #18540
- @VinciGit00 made their first contribution in #18923
Full Changelog: v1.80.13-nightly...v1.80.15-nightly