BerriAI/litellm v1.75.2-nightly

on GitHub

What's Changed

• [Redis] - Add ability to add client through GCP IAM Auth by @jugaldb in #13275
• [LLM Translation + Coding tools] Added litellm claude code count tokens support by @jugaldb in #13261
• [LLM Translation] claude opus 4.1 support for anthropic provider by @jugaldb in #13296
• [New model] add bedrock/us.anthropic.claude-opus-4-1-20250805-v1:0 by @ishaan-jaff in #13295
• New models - add fireworks_ai/glm-4p5 model family by @ishaan-jaff in #13297
• [Feat] Add fireworks gpt-oss models by @ishaan-jaff in #13303
• Create New Key - Make Team Field Required for Service Account by @NANDINI-star in #13302
• [LLM Translation] Fix model group on clientside auth with API calls by @jugaldb in #13293
• [Redis IAM] Change documentation by @jugaldb in #13306
• [Proxy server] Add apscheduler log suppress by @jugaldb in #13299
• Fix double slash issue in SSO login URL construction by @NANDINI-star in #13289
• [MCP Gateway] refactor mcp guardrails by @jugaldb in #13238
• [Bug]: Fix Mimetype Resolution Error in Bedrock Document Understanding by @ishaan-jaff in #13309
• [Feat] - When using custom tags on prometheus allow using wildcard patterns by @ishaan-jaff in #13316
• [Bug Fix] [Bug]: New Databricks Foundation Models databricks-gpt-oss-20b and databricks-gpt-oss-120b failed with error: litellm.APIConnectionError: 'signature' by @ishaan-jaff in #13318
• [LLM Translation] Fix model group on clientside auth with API calls by @jugaldb in #13314
• fix(streaming_handler.py): include cost in streaming usage object by @krrishdholakia in #13319
• feat(JinaAI): support multimodal embedding models by @zjx20 in #13181
• Exclude none fields on /chat/completion - fixes n8n bug + Allow calling /v1/models when end user over budget by @krrishdholakia in #13320
• feat: Add logo customization for LiteLLM admin UI by @colesmcintosh in #12958
• [Feat] - New model - Add Bedrock gpt oss models - "openai.gpt-oss-20b-1:0", "openai.gpt-oss-120b-1:0" by @ishaan-jaff in #13342
• [MCP Gateway] fix auth on ui for bearer servers by @jugaldb in #13312
• Fix create, search vector store error by @yeahyung in #13285
• [Fix migration for MCP server name and alias] added new migration files by @jugaldb in #13345
• [MCP Gateway] Added route check for internal users by @jugaldb in #13350
• [UI]added token breakdown in ui by @jugaldb in #13357
• Feat - New models add groq/openai/gpt-oss by @ishaan-jaff in #13363
• Provider logos on usage page by @NANDINI-star in #13372
• feat(integrations): allow setting of braintrust callback base url by @edwarddamato in #13368
• fix: 12152 - Redacted sensitive information logged in guardrails by @AnandKhinvasara in #13356
• Add labels to migrations job template by @unique-jakub in #13343
• Update OCI docs by @breno-aumo in #13336
• [Feat] Responses API Session Handling - Multi media support by @ishaan-jaff in #13347
• Add GPT 5 models by @lowjiansheng in #13377
• feat - add claude-opus-4-1 on cost map by @ishaan-jaff in #13384
• feat: Add GPT-5 model family with official OpenAI specifications (#13… by @ishaan-jaff in #13386
• [Feat] add azure/gpt-5 model family by @ishaan-jaff in #13385
• Revert "Fix SSO Logout | Create Unified Login Page with SSO and Usern… by @ishaan-jaff in #13387
• Fix non-root docker image for migration by @1995parham in #13379
• [Bug Fix] Mistral Tool Calling - Grammar error: at 3(11): failed to compile JSON schema by @ishaan-jaff in #13389
• [Bug Fix] OpenAI gpt-5 series does not support "max_tokens" parameter and temperature values that are not = 1 by @ishaan-jaff in #13390
• Add presidio MCP pre call docs by @jugaldb in #13392

New Contributors

• @zjx20 made their first contribution in #13181
• @edwarddamato made their first contribution in #13368

Full Changelog: v1.75.0-nightly...v1.75.2-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.75.2-nightly

Don't want to maintain your internal proxy? get in touch 🎉

Hosted Proxy Alpha: https://calendly.com/d/4mp-gd3-k5k/litellm-1-1-onboarding-chat

Load Test LiteLLM Proxy Results