LiteLLM Files

What's Changed

• fix gemini web search requests count by @KeremTurgutlu in https://github.com/BerriAI/litellm/pull/17921
• fix(perplexity): use API-provided cost instead of manual calculation by @Chesars in https://github.com/BerriAI/litellm/pull/17887
• feat(stability): add Stability AI image generation support by @Chesars in https://github.com/BerriAI/litellm/pull/17894
• fix(anthropic): use dynamic max_tokens based on model by @Chesars in https://github.com/BerriAI/litellm/pull/17900
• fix: pass credentials to PredictionServiceClient for Vertex AI custom endpoints by @dongbin-lunark in https://github.com/BerriAI/litellm/pull/17757
• Add Azure Cohere 4 reranking models by @emerzon in https://github.com/BerriAI/litellm/pull/17961
• add MCP auth header propagation by @uc4w6c in https://github.com/BerriAI/litellm/pull/17963
• Fix: add OpenAI-compatible API for Anthropic with modify_params=True by @Chesars in https://github.com/BerriAI/litellm/pull/17106
• fix(openai/responses/guardrail_translation): fix basemodel import by @krrishdholakia in https://github.com/BerriAI/litellm/pull/17977
• Guardrails API - support LLM tool call response checks on /chat/completions, /v1/responses, /v1/messages on regular + streaming calls by @krrishdholakia in https://github.com/BerriAI/litellm/pull/17619
• OpenRouter GPT 5.2, Mistral 3, and Devstral 2 by @SamAcctX in https://github.com/BerriAI/litellm/pull/17844
• add: litellm team PR template by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/17983
• Add: CI/CD rules to default PR template for LiteLLM team. by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/17985
• fix: cost calculation of gpt-image 1 model by @Sameerlite in https://github.com/BerriAI/litellm/pull/17966
• Add support for reasoning param for fireworks AI models by @Sameerlite in https://github.com/BerriAI/litellm/pull/17967
• Add provider specific tools support in responses api by @Sameerlite in https://github.com/BerriAI/litellm/pull/17980
• [Refactor] lazy imports: Use per-attribute lazy imports and extract shared constants by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/17994
• [Refactor] litellm/init.py: lazy load http handlers by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/17997
• [Refactor] litellm/init.py: lazy load caches by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18001
• [Refactor] litellm/init.py: lazy load get_modified_max_tokens by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18002
• [docs] update SAP docs by @vasilisazayka in https://github.com/BerriAI/litellm/pull/17974
• [Feat] Guardrails - litellm content filter by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18007
• feat(custom_llm): add image_edit and aimage_edit support by @Chesars in https://github.com/BerriAI/litellm/pull/17999
• fix: mcp deepcopy error by @uc4w6c in https://github.com/BerriAI/litellm/pull/18010
• [Feat] New provider - Agent Gateway, add pydantic ai agents by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18013
• fix(anthropic): claude-3-7-sonnet max_tokens to 64K default by @Chesars in https://github.com/BerriAI/litellm/pull/17979
• [fix] add qwen3-embedding-8b input per token price by @shivamrawat1 in https://github.com/BerriAI/litellm/pull/18018
• fix(gemini): use JSON instead of form-data for image edit requests by @Chesars in https://github.com/BerriAI/litellm/pull/18012
• Daily litellm staging branch by @krrishdholakia in https://github.com/BerriAI/litellm/pull/18020
• feat(gemini): support extra_headers in batch embeddings by @qdrddr in https://github.com/BerriAI/litellm/pull/18004
• Propagate token usage when generating images with Gemini by @komarovd95 in https://github.com/BerriAI/litellm/pull/17987
• feat(venice.ai): add support for Venice.ai API via providers.json by @donicrosby in https://github.com/BerriAI/litellm/pull/17962
• Litellm bedrock guardrails block precedence over masking by @kothamah in https://github.com/BerriAI/litellm/pull/17968
• Revert "Litellm bedrock guardrails block precedence over masking" by @krrishdholakia in https://github.com/BerriAI/litellm/pull/18022
• Revert "Revert "Litellm bedrock guardrails block precedence over masking"" by @krrishdholakia in https://github.com/BerriAI/litellm/pull/18023
• Fix get_model_from_request() to extract model ID from Vertex AI passthrough URLs by @krisxia0506 in https://github.com/BerriAI/litellm/pull/17970
• [Feat] New Provider - VertexAI Agent Engine by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18014
• feat(pillar): add masking support and MCP call support by @eagle-p in https://github.com/BerriAI/litellm/pull/17959
• fix: Support Signed URLs with Query Parameters in Image Processing by @OlivverX in https://github.com/BerriAI/litellm/pull/17976
• [Docs] Add docs on using pydantic ai agents with LiteLLM A2a gateway by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18026
• chore: improve issue labeling with component dropdown and more provider keywords by @Chesars in https://github.com/BerriAI/litellm/pull/17957
• Cleanup PR template: remove redundant fields by @Chesars in https://github.com/BerriAI/litellm/pull/17956
• Added new step into rotate master key function for processing credentials table by @Eric84626 in https://github.com/BerriAI/litellm/pull/17952
• [Docs] Litellm add docs vertex ai engine by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18027
• Litellm dev 12 15 2025 p1 by @krrishdholakia in https://github.com/BerriAI/litellm/pull/18028
• [Feature] UI - Milvus Vector Store by @yuneng-jiang in https://github.com/BerriAI/litellm/pull/18030
• fix: add headers to metadata for guardrails on pass-through endpoints by @NicolaivdSmagt in https://github.com/BerriAI/litellm/pull/17992
• Router order parameter documentation by @krrishdholakia in https://github.com/BerriAI/litellm/pull/18045
• [Refactor] litellm/init.py: lazy load LLMClientCache by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18008
• [Refactor] litellm/init.py: lazy load bedrock types by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18053
• [Refactor] litellm/init.py: lazy load .types.utils by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18054
• [Refactor] litellm/init.py: lazy load dotprompt integration by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18056
• [Refactor] litellm/init.py: lazy load default encoding from client decorator by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18059
• [Feature] Download Prisma binaries at build time instead of at runtime for Security Restricted environments by @mdiloreto in https://github.com/BerriAI/litellm/pull/17695
• Add custom headers in responses API by @Sameerlite in https://github.com/BerriAI/litellm/pull/18036
• fix: skip adding beta headers for vertex ai as it is not suppported by @Sameerlite in https://github.com/BerriAI/litellm/pull/18037
• Remove ttl field when routing to bedrock by @Sameerlite in https://github.com/BerriAI/litellm/pull/18049
• fix: Add none to encoding_format instead of omitting it by @Sameerlite in https://github.com/BerriAI/litellm/pull/18042
• Add support for agent skills in chat completion by @Sameerlite in https://github.com/BerriAI/litellm/pull/18031
• Fix managed files endpoint by @Sameerlite in https://github.com/BerriAI/litellm/pull/18046
• Revert "Fix get_model_from_request() to extract model ID from Vertex AI passthrough URLs" by @Sameerlite in https://github.com/BerriAI/litellm/pull/18063
• [Refactor] litellm/init.py: lazy-load heavy client decorator imports by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18064
• Litellm staging 12 16 2025 by @krrishdholakia in https://github.com/BerriAI/litellm/pull/18025
• [Refactor] litellm/init.py: lazy-load heavy imports from litellm.main by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18066
• [Refactor] litellm/init.py: lazy-load AmazonConverseConfig by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18069
• [Refactor] litellm/init.py: lazy load encoding from main.py by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18070
• [Feature] UI - Add Models Conditional Rendering by @yuneng-jiang in https://github.com/BerriAI/litellm/pull/18071
• [Refactor] litellm/init.py: lazy load GuardrailItem by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18072
• [Refactor] images/main.py: lazy load ImageEditRequestUtils by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18074
• Lazy load OpenAILikeChatConfig to avoid heavy import by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/18075
• [Feat] LiteLLM Content Filter - Add Support for Brazil PII field by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18076
• Add Azure DeepSeek V3.2 versions by @emerzon in https://github.com/BerriAI/litellm/pull/18019
• feat: add github_copilot model info by @codgician in https://github.com/BerriAI/litellm/pull/17858
• [Feat] New Endpoint - Google Interactions API - added on litellm SDK by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18079
• [Feat] Add New Google Interactions API on AI Gateway by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/18081
• Allow base_model for non Azure providers in proxy by @jyeros in https://github.com/BerriAI/litellm/pull/18038
• docs: add documentation describing configurable Hashicorp Vault mount… by @uc4w6c in https://github.com/BerriAI/litellm/pull/18082
• [Feature] Add LiteLLM Overhead to Logs by @yuneng-jiang in https://github.com/BerriAI/litellm/pull/18033
• [Feature] UI - Show LiteLLM Overhead in Logs by @yuneng-jiang in https://github.com/BerriAI/litellm/pull/18034

New Contributors

• @dongbin-lunark made their first contribution in https://github.com/BerriAI/litellm/pull/17757
• @qdrddr made their first contribution in https://github.com/BerriAI/litellm/pull/18004
• @donicrosby made their first contribution in https://github.com/BerriAI/litellm/pull/17962
• @NicolaivdSmagt made their first contribution in https://github.com/BerriAI/litellm/pull/17992

Full Changelog: https://github.com/BerriAI/litellm/compare/v1.80.10-nightly...v1.80.10.rc.3