41 Releases published by 2 people

• v1.74.2-nightly

  published Jul 11, 2025

• v1.74.3-nightly

  published Jul 12, 2025

• v1.74.3-stable-draft

  published Jul 12, 2025

• v1.74.3.rc.1

  published Jul 13, 2025

• v1.74.3.rc.2

  published Jul 13, 2025

• v1.74.3.rc.3

  published Jul 16, 2025

• v1.74.3.dev2

  published Jul 16, 2025

• v1.74.4-nightly

  published Jul 17, 2025

• v1.74.3-stable

  published Jul 18, 2025

• v1.74.5.dev1

  published Jul 18, 2025

• v1.74.6-nightly

  published Jul 19, 2025

• v1.74.7-nightly

  published Jul 20, 2025

• v1.74.3-stable.patch.1

  published Jul 21, 2025

• v1.74.3-stable.patch.2

  published Jul 21, 2025

• v1.74.3-stable.patch.3

  published Jul 21, 2025

• v1.74.7.rc.1

  published Jul 21, 2025

• v1.72.2-stable.debug

  published Jul 22, 2025

• v1.74.3-stable.patch.4

  published Jul 23, 2025

• v1.74.8-nightly

  published Jul 24, 2025

• v1.74.7-stable

  published Jul 25, 2025

• v1.74.7-stable.patch.1

  published Jul 25, 2025

• v1.74.7-stable.patch.2

  published Jul 25, 2025

• v1.74.9.rc-draft

  published Jul 26, 2025

• litellm_v1.65.4-dev_fix

  published Jul 28, 2025

• v1.74.9.rc.1

  published Jul 29, 2025

• v1.74.12-nightly

  published Jul 31, 2025

• v1.74.9-stable

  published Aug 1, 2025

• v1.74.14-nightly

  published Aug 2, 2025

• v1.74.14.dev1

  published Aug 2, 2025

• v1.74.15-nightly

  published Aug 2, 2025

• 1.74.15.rc.1

  published Aug 2, 2025

• v1.74.9-stable.patch.1

  published Aug 3, 2025

• v1.74.15.rc.2

  published Aug 5, 2025

• v1.75.0-nightly

  published Aug 5, 2025

• v1.75.0.dev2

  published Aug 5, 2025

• v1.75.2-nightly

  published Aug 8, 2025

• v1.75.3-nightly

  published Aug 8, 2025

• v1.74.15-stable

  published Aug 8, 2025

• v1.75.4-nightly

  published Aug 9, 2025

• v1.75.5-stable.rc-draft

  published Aug 9, 2025

• v1.75.5.rc.1

  published Aug 10, 2025

435 Pull requests merged by 86 people

• [Bug Fix] - Allow using reasoning_effort for gpt-5 model family and reasoning for Responses API

  #13475 merged Aug 10, 2025

• ui - build updated ui + increase max_tokens in health_check = 10 (gpt-5-nano throws error for max_token=1)

  #13482 merged Aug 10, 2025

• Litellm model cost map fixes

  #13480 merged Aug 10, 2025

• Litellm release notes 08 10 2025

  #13479 merged Aug 10, 2025

• Model Hub - show price for azure models + Responses API - support custom tool

  #13418 merged Aug 9, 2025

• Add digitalocean provider

  #12169 merged Aug 9, 2025

• [Bug]: Fix JWTs access not working with model groups

  #13474 merged Aug 9, 2025

• [MCP Gateway] Add local storage auth for tokens

  #13473 merged Aug 9, 2025

• [Proxy changes] Litellm add model price reload schedule for multi-pod

  #13470 merged Aug 9, 2025

• Router - reduce p99 latency w/ redis enabled by 50% + OTEL - track pre_call hook latency

  #13362 merged Aug 9, 2025

• Latency Improvements: only track tpm/rpm usage when set on deployment+ LLM Caching - use an in-memory cache to reduce redis calls + OTEL - track time spent on LLM caching

  #13472 merged Aug 9, 2025

• [Bug Fix] Allow using Swagger for /chat/completions

  #13469 merged Aug 9, 2025

• [Proxy + UI] Litellm add reload model api and button

  #13464 merged Aug 9, 2025

• [LLM Translation] Litellm azure o series drop params

  #13353 merged Aug 9, 2025

• [Bug Fix] /health checks should not use hardcoded trace_id

  #13468 merged Aug 9, 2025

• [Bug Fix] - Get Routes

  #13466 merged Aug 9, 2025

• [Bug Fix] Responses API - Responses API failed if input containing ResponseReasoningItem

  #13465 merged Aug 9, 2025

• docs - native LiteLLM prompt mgmt

  #13463 merged Aug 9, 2025

• feat(models): add OpenRouter and Cerebras GPT-OSS (20b, 120b) to cost map

  #13442 merged Aug 9, 2025

• [Documentation ]added mcp guardrails doc in mcp.md

  #13452 merged Aug 9, 2025

• [Feat] Working e2e flow for Responses API session management with media

  #13456 merged Aug 9, 2025

• feat(reasoning): support 'minimal' effort type for OpenAI

  #13447 merged Aug 9, 2025

• Disable logging settings for non-enterprise users - Create Key

  #13431 merged Aug 9, 2025

• LiteLLM UI - Test Key Page - allow uploading images for /chat/completions and /responses

  #13445 merged Aug 8, 2025

• [Docs] Add docs on how router / cooldowns work

  #13444 merged Aug 8, 2025

• [Bug Fix] Improve error message from - key creation permission error message

  #13443 merged Aug 8, 2025

• Add support for reasoning_effort minimal

  #13401 merged Aug 8, 2025

• LLM Translation - fix prices for oai gpt 5

  #13441 merged Aug 8, 2025

• [Bug fix] - Error creating standard logging object - can't register atexit after shutdownLitellm fixes standard logging payload

  #13436 merged Aug 8, 2025

• Display Error from Backend on the UI - Notification

  #13427 merged Aug 8, 2025

• [User Delete from team] fix user membership issue

  #13433 merged Aug 8, 2025

• [Feat] Add reasoning_effort to OpenAIGPT5Config

  #13434 merged Aug 8, 2025

• fix(proxy): add missing braintrust api base to env vars

  #13412 merged Aug 8, 2025

• fix(access group): allow access group on mcp tool retrieval

  #13425 merged Aug 8, 2025

• Correct GPT-5 token limits and price

  #13423 merged Aug 8, 2025

• feat(usage): Usage - Ensure Usage Page loads after the DB has large entries (aggregated user daily activity endpoint and UI integrati…

  #13400 merged Aug 8, 2025

• feat(usage): aggregated user daily activity endpoint and UI integration

  #13395 merged Aug 8, 2025

• Add Custom Tooltips to Model Mapping Table

  #13294 merged Aug 8, 2025

• [Bug Fix] Responses api session management for streaming responses

  #13396 merged Aug 8, 2025

• fix(responses api): fix streaming ID consistency and tool format handling

  #12640 merged Aug 8, 2025

• Add presidio MCP pre call docs

  #13392 merged Aug 7, 2025

• [Bug Fix] OpenAI gpt-5 series does not support "max_tokens" parameter and temperature values that are not = 1

  #13390 merged Aug 7, 2025

• [Bug Fix] Mistral Tool Calling - Grammar error: at 3(11): failed to compile JSON schema

  #13389 merged Aug 7, 2025

• Fix non-root docker image for migration

  #13379 merged Aug 7, 2025

• Revert "Fix SSO Logout | Create Unified Login Page with SSO and Usern…

  #13387 merged Aug 7, 2025

• [Feat] add azure/gpt-5 model family

  #13385 merged Aug 7, 2025

• feat: Add GPT-5 model family with official OpenAI specifications (#13…

  #13386 merged Aug 7, 2025

• feat: Add GPT-5 model family with official OpenAI specifications

  #13378 merged Aug 7, 2025

• feat - add claude-opus-4-1 on cost map

  #13384 merged Aug 7, 2025

• Add GPT 5 models

  #13377 merged Aug 7, 2025

• [Feat] Responses API Session Handling - Multi media support

  #13347 merged Aug 7, 2025

• Update OCI docs

  #13336 merged Aug 7, 2025

• Add labels to migrations job template

  #13343 merged Aug 7, 2025

• fix: 12152 - Redacted sensitive information logged in guardrails

  #13356 merged Aug 7, 2025

• feat(integrations): allow setting of braintrust callback base url

  #13368 merged Aug 7, 2025

• Provider logos on usage page

  #13372 merged Aug 7, 2025

• Feat - New models add groq/openai/gpt-oss

  #13363 merged Aug 7, 2025

• [UI]added token breakdown in ui

  #13357 merged Aug 7, 2025

• [MCP Gateway] Added route check for internal users

  #13350 merged Aug 6, 2025

• [Fix migration for MCP server name and alias] added new migration files

  #13345 merged Aug 6, 2025

• Fix create, search vector store error

  #13285 merged Aug 6, 2025

• [MCP Gateway] fix auth on ui for bearer servers

  #13312 merged Aug 6, 2025

• [Feat] - New model - Add Bedrock gpt oss models - "openai.gpt-oss-20b-1:0", "openai.gpt-oss-120b-1:0"

  #13342 merged Aug 6, 2025

• feat: Add logo customization for LiteLLM admin UI

  #12958 merged Aug 6, 2025

• Exclude none fields on /chat/completion - fixes n8n bug + Allow calling /v1/models when end user over budget

  #13320 merged Aug 6, 2025

• feat(JinaAI): support multimodal embedding models

  #13181 merged Aug 6, 2025

• fix(streaming_handler.py): include cost in streaming usage object

  #13319 merged Aug 6, 2025

• [LLM Translation] Fix model group on clientside auth with API calls

  #13314 merged Aug 6, 2025

• [Bug Fix] [Bug]: New Databricks Foundation Models databricks-gpt-oss-20b and databricks-gpt-oss-120b failed with error: litellm.APIConnectionError: 'signature'

  #13318 merged Aug 6, 2025

• [Feat] - When using custom tags on prometheus allow using wildcard patterns

  #13316 merged Aug 6, 2025

• [Bug]: Fix Mimetype Resolution Error in Bedrock Document Understanding

  #13309 merged Aug 6, 2025

• Litellm fix OpenAI spec tools

  #13315 merged Aug 6, 2025

• [MCP Gateway] refactor mcp guardrails

  #13238 merged Aug 5, 2025

• Fix double slash issue in SSO login URL construction

  #13289 merged Aug 5, 2025

• [Proxy server] Add apscheduler log suppress

  #13299 merged Aug 5, 2025

• [Redis IAM] Change documentation

  #13306 merged Aug 5, 2025

• [LLM Translation] Fix model group on clientside auth with API calls

  #13293 merged Aug 5, 2025

• Create New Key - Make Team Field Required for Service Account

  #13302 merged Aug 5, 2025

• [Feat] Add fireworks gpt-oss models

  #13303 merged Aug 5, 2025

• New models - add fireworks_ai/glm-4p5 model family

  #13297 merged Aug 5, 2025

• [New model] add bedrock/us.anthropic.claude-opus-4-1-20250805-v1:0

  #13295 merged Aug 5, 2025

• [LLM Translation] claude opus 4.1 support for anthropic provider

  #13296 merged Aug 5, 2025

• [LLM Translation + Coding tools] Added litellm claude code count tokens support

  #13261 merged Aug 5, 2025

• [Redis] - Add ability to add client through GCP IAM Auth

  #13275 merged Aug 5, 2025

• Revert "Fix: Langfuse reporting "client closed" error due to httpx client TTL"

  #13291 merged Aug 5, 2025

• fix(main.py): handle tool being a pydantic object + Fix unpack defs deepcopy issue for bedrock

  #13274 merged Aug 5, 2025

• Ensure disable_llm_api_endpoints works + Add wildcard model support for 'team-byok' model

  #13278 merged Aug 5, 2025

• fix OCI linting errors

  #13279 merged Aug 5, 2025

• [UI] - Add ability to set model alias per key/team

  #13276 merged Aug 5, 2025

• [LLM Translation] Support /v1/models/{model_id} retrieval

  #13268 merged Aug 5, 2025

• [LLM Translation] input cost per token higher than $1 test

  #13270 merged Aug 5, 2025

• [Bug Fix] Prometheus - fix for litellm_input_tokens_metric, litellm_output_tokens_metric - Note this updates the metric name

  #13271 merged Aug 5, 2025

• [LLM Translation] Correct pricing for web search on 4o-mini

  #13269 merged Aug 5, 2025

• Fix: Langfuse reporting "client closed" error due to httpx client TTL

  #13045 merged Aug 4, 2025

• Add GCS bucket caching support

  #13122 merged Aug 4, 2025

• Support OCI provider

  #13206 merged Aug 4, 2025

• [Bug Fix] Fix Server root path regression on UI when using "Login"

  #13267 merged Aug 4, 2025

• Minor formatting changes to token-cost.json

  #13244 merged Aug 4, 2025

• Bug Fix - Responses API raises error with Gemini Tool Calls in input

  #13260 merged Aug 4, 2025

• [Bug Fix] OpenAI / Azure Responses API - Add service_tier , safety_identifier supported params

  #13258 merged Aug 4, 2025

• [UI] Add team deletion check for teams with keys

  #12953 merged Aug 4, 2025

• [LLM Translation] Fix Model Usage not having text tokens

  #13234 merged Aug 4, 2025

• [Proxy] Add OpenShift Support to non root docker image

  #13239 merged Aug 3, 2025

• Prompt Management - add prompts on UI

  #13240 merged Aug 3, 2025

• Prompt Management - Add table + prompt info page to UI

  #13232 merged Aug 3, 2025

• UI - Add giving keys prompt access

  #13233 merged Aug 3, 2025

• [docs release notes]

  #13237 merged Aug 2, 2025

• [UI QA Fixes] Stable release

  #13231 merged Aug 2, 2025

• Prompt Management (2/2) - New /prompt/list endpoint + key-based access to prompt templates

  #13218 merged Aug 2, 2025

• Revert "fix: role chaining and session name with webauthentication for aws bedrock"

  #13230 merged Aug 2, 2025

• [QA Fixes for MCP] - Ensure MCPs load + don't run a health check everytime we load MCPs on UI

  #13228 merged Aug 2, 2025

• Prompt Management - abstract prompt templates away from model list (enables permission management on prompt templates)

  #13219 merged Aug 2, 2025

• litellm/proxy: preserve model order of /v1/models and /model_group/info

  #13178 merged Aug 2, 2025

• Fix missing extra_headers support for vLLM/openai_like embeddings

  #13198 merged Aug 2, 2025

• fix: role chaining and session name with webauthentication for aws bedrock

  #13205 merged Aug 2, 2025

• Add Perplexity citation annotations support

  #13225 merged Aug 2, 2025

• Add advanced date picker to all the tabs on the usage page

  #13221 merged Aug 2, 2025

• [MCP Gateway] Litellm mcp pre and during guardrails

  #13188 merged Aug 2, 2025

• [LLM] - suppress httpx logging

  #13217 merged Aug 2, 2025

• [LLM] fix model reload on model update

  #13216 merged Aug 2, 2025

• [Proxy]fix key mgmt

  #13148 merged Aug 2, 2025

• [Separate Health App] Update Helm Deployment.yaml

  #13162 merged Aug 1, 2025

• [QA] Viewing Agent Activity Headers on UI Usage Page

  #13212 merged Aug 1, 2025

• [LLM translation] Fix bedrock computer use #13143

  #13150 merged Aug 1, 2025

• Index.md - cleanup docs

  #13215 merged Aug 1, 2025

• Fix API Key Being Logged

  #12978 merged Aug 1, 2025

• Allow to redifine LLM base api URL in the pass trough endpoints

  #13134 merged Aug 1, 2025

• add openssl in apk install in runtime stage in dockerfile.non_root

  #13168 merged Aug 1, 2025

• feat(helm): allow helm hooks for migrations job

  #13174 merged Aug 1, 2025

• Fix/panw prisma airs post call hook

  #13185 merged Aug 1, 2025

• [UI QA] QA - Agent Activity Tab

  #13203 merged Aug 1, 2025

• Anthropic - mid stream fallbacks p2 (add token usage across both calls)

  #13170 merged Aug 1, 2025

• Anthropic - working mid-stream fallbacks

  #13149 merged Aug 1, 2025

• Fix - using managed files w/ OTEL + UI - add model group alias on UI

  #13171 merged Aug 1, 2025

• [Docs] Add details on when to use specific health endpoints

  #13193 merged Aug 1, 2025

• Fix langfuse test patch path causing CI failures

  #13192 merged Aug 1, 2025

• Litellm fix fallbacks UI

  #13191 merged Aug 1, 2025

• [Feat] Allow redacting message / response content for specific logging integrations - DD LLM Observability

  #13158 merged Jul 31, 2025

• [Bug Fix] Infra - ensure that stale Prisma clients disconnect DB connection

  #13140 merged Jul 31, 2025

• [Bug Fix] Gemini-CLI Integration - ensure tool calling works as expected on generateContent

  #13189 merged Jul 31, 2025

• fix: support negative indexes in cache_control_injection_points for Anthropic Claude (#10226)

  #13187 merged Jul 31, 2025

• [Proxy Startup]fix db config through envs

  #13111 merged Jul 31, 2025

• [Feat] Background Health Checks - Allow disabling background health checks for a specific

  #13186 merged Jul 31, 2025

• fix: remove obsolete attribute version in docker compose

  #13172 merged Jul 31, 2025

• add framework name to UserAgent header in AWS Bedrock API call

  #13159 merged Jul 31, 2025

• build(config.yml): migrate build_and_test to ci/cd pg db

  #13166 merged Jul 31, 2025

• [MCP Gateway] fix migrations

  #13157 merged Jul 30, 2025

• [MCP Gateway] Litellm mcp client list fail

  #13114 merged Jul 30, 2025

• Litellm explore postgres db ci cd

  #13156 merged Jul 30, 2025

• [Feat] v2 updates - tracking DAU, WAU, MAU for coding tool usage + show Daily Usage per User

  #13147 merged Jul 30, 2025

• [MCP Guardrails] move pre and during hooks to ProxyLoggin

  #13109 merged Jul 30, 2025

• [LLM translation] Fix bedrock computer use

  #13143 merged Jul 30, 2025

• [Feat] UI + Backend add a tab for use agent activity

  #13146 merged Jul 30, 2025

• New Advanced Date Range Picker Component

  #13141 merged Jul 30, 2025

• [Proxy UI] fix object permission for orgs

  #13142 merged Jul 30, 2025

• Resolve Anthropic Overloaded error during stream

  #9809 merged Jul 30, 2025

• Added Voyage, Jinai, Deepinfra and VolcEngine providers on the UI

  #13131 merged Jul 30, 2025

• [MCP Protocol header] fix issue with clients protocol header

  #13112 merged Jul 30, 2025

• [MCP Gateway] add health check endpoints for MCP

  #13106 merged Jul 30, 2025

• fix tool aws bedrock call index when the function only have optional arg

  #13115 merged Jul 30, 2025

• move to use_prisma_migrate by default + resolve team-only models on auth checks + UI - add sagemaker on UI

  #13117 merged Jul 30, 2025

• fix(model_checks.py): handle custom values in wildcard model name (e.g. genai/test/*)

  #13116 merged Jul 30, 2025

• Revert "[LLM translation] Add support for bedrock computer use"

  #13118 merged Jul 30, 2025

• After selecting date range show loader on usage cost charts

  #13113 merged Jul 30, 2025

• [LLM translation] Add support for bedrock computer use

  #12948 merged Jul 29, 2025

• [Feat] MLFlow Logging - Allow adding tags for ML Flow logging requests

  #13108 merged Jul 29, 2025

• feat: Add dot notation support for all JWT fields

  #13013 merged Jul 29, 2025

• Custom Auth - bubble up custom exceptions

  #13093 merged Jul 29, 2025

• Fix token counter to ignore unsupported keys like prefix (#11791)

  #11954 merged Jul 29, 2025

• [MCP Gateway] Add protocol headers

  #13062 merged Jul 29, 2025

• Fix/gemini api key environment variable support

  #12507 merged Jul 29, 2025

• Fix fallback delete

  #12606 merged Jul 29, 2025

• Revert "[Bug]: Set user from token user_id for OpenMeter integration"

  #13107 merged Jul 29, 2025

• [Bug]: Set user from token user_id for OpenMeter integration

  #13029 merged Jul 29, 2025

• fix: helm migration job not running schema update

  #12809 merged Jul 29, 2025

• BUGFIX: Jitter should be added not multiplied (#12877)

  #12901 merged Jul 29, 2025

• Optionally enable GenAI telemetry following semantic conventions

  #12626 merged Jul 29, 2025

• fix: always use choice index=0 for Anthropic streaming responses

  #12666 merged Jul 29, 2025

• [Infra] Looses MCP python version restrictions

  #13102 merged Jul 29, 2025

• [LLM translation] add openrouter grok4

  #13018 merged Jul 29, 2025

• [Feat] Allow using query_params for setting API Key for generateContent routes

  #13100 merged Jul 29, 2025

• set default value for mcp namespace tool name in spend table to prevent duplicate entry in table

  #12894 merged Jul 29, 2025

• [Bug Fix] Gemini-CLI - The Gemini Custom API request has an incorrect authorization format

  #13098 merged Jul 29, 2025

• fix: improve MCP server URL validation to support internal/Kubernetes URLs

  #13099 merged Jul 29, 2025

• [MCP gateway] add pre and during call hooks init

  #13067 merged Jul 29, 2025

• Fix list team v2 security check

  #13094 merged Jul 29, 2025

• Remove extraneous s in docs

  #13079 merged Jul 29, 2025

• Azure api_version="preview" support + Bedrock cost tracking via Anthropic /v1/messages

  #13072 merged Jul 29, 2025

• Fix anthropic passthrough logging handler model fallback for streaming requests

  #13022 merged Jul 29, 2025

• docs: add Qwen Code CLI tutorial

  #12915 merged Jul 29, 2025

• Added handling for pwd protected cert files in AOAI CertificateCreden…

  #12995 merged Jul 29, 2025

• Default Usage Chart Date Range: Last 7 Days

  #12917 merged Jul 29, 2025

• [Feat] Add Google AI Studio Imagen4 model family

  #13065 merged Jul 29, 2025

• [Bug Fix] The model gemini-2.5-flash with the merge_reasoning_content_in_choices parameter does not work

  #13066 merged Jul 29, 2025

• [MCP gateway] add url namespacing docs

  #13063 merged Jul 29, 2025

• [MCP Gateway] MCP tools fix scrolling issue

  #13015 merged Jul 29, 2025

• feat(langfuse-otel): Add comprehensive metadata support to Langfuse OpenTelemetry integration

  #12956 merged Jul 28, 2025

• chore: Improve docs for cost tracking

  #12976 merged Jul 28, 2025

• fix: correct CompletionRequest messages type to match OpenAI API spec

  #12980 merged Jul 28, 2025

• Properly parse json options for key generation in the UI

  #12989 merged Jul 28, 2025

• Remove duplicate test case verifying field filtering logic

  #13023 merged Jul 28, 2025

• [Bug Fix] Pass through logging handler VertexAI - ensure multimodal embedding responses are logged

  #13050 merged Jul 28, 2025

• docs - openwebui show how to include reasoning content for gemini models

  #13060 merged Jul 28, 2025

• add X-Initiator header for GitHub Copilot to reduce premium requests

  #13016 merged Jul 28, 2025

• Bulk User Edit - additional improvements - edit all users + set 'no-default-models' on all users

  #12925 merged Jul 27, 2025

• VertexAI - camelcase optional params for image generation + Anthropic - streaming, always ensure assistant role set on only first chunk

  #12889 merged Jul 27, 2025

• Litellm release notes 07 27 2025 p1

  #13027 merged Jul 27, 2025

• [MCP Gateway] Litellm mcp multi header propagation

  #13003 merged Jul 26, 2025

• UI SSO - fix reset env var when ui_access_mode is updated

  #13011 merged Jul 26, 2025

• [FEAT] Model-Guardrails: Add on UI

  #13006 merged Jul 26, 2025

• [Vector Store] make vector store permission management OSS

  #12990 merged Jul 26, 2025

• Fixup ollama model listing (again)

  #13008 merged Jul 26, 2025

• [MCP Gateway] add Litellm mcp alias for prefixing

  #12994 merged Jul 26, 2025

• [BUG Fix] Cannot pickle coroutine object

  #13005 merged Jul 26, 2025

• Fix issue writing db

  #13001 merged Jul 26, 2025

• build: update pip package

  #12998 merged Jul 25, 2025

• [MCP Gateway] Move cost tracking and permission management to OSS

  #12988 merged Jul 25, 2025

• [LLM Translation] fix query params for realtime api intent

  #12838 merged Jul 25, 2025

• clean and verify key before inserting

  #12840 merged Jul 25, 2025

• [LLM Translation] Add bytedance/ui-tars-1.5-7b on openrouter

  #12882 merged Jul 25, 2025

• build(github/manual_pypi_publish.yml): manual workflow to publish pip package - used for pushing dev releases

  #12985 merged Jul 25, 2025

• Guardrails - support model-level guardrails

  #12968 merged Jul 25, 2025

• Show global retry policy on UI

  #12969 merged Jul 25, 2025

• GuardrailsAI: use validatedOutput to allow usage of "fix" guards

  #12891 merged Jul 25, 2025

• fix(auth_utils): make header comparison case-insensitive

  #12950 merged Jul 25, 2025

• feat: add openrouter/qwen/qwen3-coder model configuration

  #12910 merged Jul 25, 2025

• Fix: Shorten Gemini tool_call_id for Open AI compatibility

  #12941 merged Jul 25, 2025

• docs: added documentation about metadata exposed over the /v1/models endpoint

  #12942 merged Jul 25, 2025

• [Feat] Add inpainting support and corresponding tests for Amazon Nova…

  #12949 merged Jul 25, 2025

• [Feat] Edit Auto Router Settings on UI

  #12966 merged Jul 25, 2025

• [Feat] UI - Allow Adding LiteLLM Auto Router on UI

  #12960 merged Jul 25, 2025

• [LLM Translation] added new realtime model for openai

  #12946 merged Jul 25, 2025

• [LLM Translation] - Bug fix Anthropic Tool calling

  #12959 merged Jul 25, 2025

• [Feat] Backend Router - Add Auto-Router powered by semantic-router

  #12955 merged Jul 25, 2025

• fix(internal_user_endpoints.py): delete member from team table on /user/delete

  #12926 merged Jul 24, 2025

• Proxy - specify key_type - allows specifying if key can call LLM API routes vs. Management routes only

  #12909 merged Jul 24, 2025

• Prometheus - tags, fix '[tag]="false"' when tag is set

  #12916 merged Jul 24, 2025

• Update control_plane_and_data_plane.md

  #12939 merged Jul 24, 2025

• [UI] Allow setting up CloudZero Usage through LiteLLM UI

  #12923 merged Jul 24, 2025

• Add GA version of gemini 2.5 flash lite for both vertex and gemini

  #12920 merged Jul 24, 2025

• [Feat] LiteLLM CloudZero Integration updates - using LiteLLM_SpendLogs Table

  #12922 merged Jul 24, 2025

• [Feat] LiteLLM x Cloudzero integration - Allow exporting spend to cloudzero

  #12908 merged Jul 23, 2025

• Swagger - bump DOMPurify version + add back Support local swagger bundle (good for airgapped environments) + support local swagger on custom root paths

  #12911 merged Jul 23, 2025

• feat: Add Pillar Security guardrail integration

  #12791 merged Jul 23, 2025

• feat: extended /v1/models endpoint, now it returns with fallbacks on demand

  #12811 merged Jul 23, 2025

• rm retired anthropic models from model_prices_and_context_window.json

  #12864 merged Jul 23, 2025

• [Add health check] add architecture diagram

  #12879 merged Jul 23, 2025

• [Docs ]Litellm mcp access group doc

  #12883 merged Jul 23, 2025

• Request Headers - support x-litellm-num-retries + Usage - support usage by model group

  #12890 merged Jul 23, 2025

• Fix async retryer on .acompletion() + forward clientside headers - filter out content-type from clientside request (causes llm api call to hang)

  #12886 merged Jul 23, 2025

• [Feat] - Track cost + add tags for health checks done by LiteLLM Proxy

  #12880 merged Jul 23, 2025

• [Feat] Add cost tracking for new vertex_ai/llama-3 API models

  #12878 merged Jul 22, 2025

• [LLM Translation] Litellm gemini 2.0 live support

  #12839 merged Jul 22, 2025

• [Feat] Add Recraft API - Image Edits Support

  #12874 merged Jul 22, 2025

• [Separate Health App] Pass through cmd args via supervisord

  #12871 merged Jul 22, 2025

• Bug fix - Azure KeyVault not in image, add azure-keyvault==4.2.0 to Docker img

  #12873 merged Jul 22, 2025

• build(deps): bump form-data from 4.0.3 to 4.0.4 in /docs/my-website

  #12867 merged Jul 22, 2025

• Replace non-root Dockerfile base with Alpine multi-stage build;

  #12707 merged Jul 22, 2025

• Improvements on the Regenerate Key Flow

  #12788 merged Jul 22, 2025

• Openrouter - filter out cache_control flag for non-anthropic models (allows usage with claude code)

  #12850 merged Jul 22, 2025

• Fix async retryer on .acompletion() + forward clientside headers - filter out content-type from clientside request (causes llm api call to hang)

  #12848 merged Jul 22, 2025

• Fix team_member_budget update logic

  #12843 merged Jul 22, 2025

• build(deps): bump form-data from 4.0.0 to 4.0.4 in /ui/litellm-dashboard

  #12851 merged Jul 22, 2025

• Passthrough Auth - make Auth checks OSS + Anthropic - only show 'reasoning_effort' for supported models

  #12847 merged Jul 22, 2025

• Litellm batch cost tracking debug

  #12782 merged Jul 22, 2025

• feat: add Hyperbolic provider support

  #12826 merged Jul 22, 2025

• fix(watsonx): IBM Watsonx - use correct parameter name for tool choice

  #9980 merged Jul 22, 2025

• Docs - litellm benchmarks

  #12842 merged Jul 22, 2025

• [Azure OpenAI Feature] - Support DefaultAzureCredential without hard-coded environment variables

  #12841 merged Jul 22, 2025

• [LLM Translation] add qwen-vl-plus

  #12829 merged Jul 21, 2025

• [Feat] Add fireworks - fireworks/models/kimi-k2-instruct

  #12837 merged Jul 21, 2025

• [Bug Fix] - gemini leaking FD for sync calls with litellm.completion

  #12824 merged Jul 21, 2025

• [Feat] Add Recraft Image Generation API Support - New LLM Provider

  #12832 merged Jul 21, 2025

• Add Google Cloud Model Armor guardrail documentation

  #12814 merged Jul 21, 2025

• fix: remove deprecated groq/qwen-qwq-32b and add qwen/qwen3-32b

  #12831 merged Jul 21, 2025

• feat: add Morph provider support

  #12821 merged Jul 21, 2025

• [LLM Translation - GH Copilot] added dynamic endpoint support

  #12827 merged Jul 21, 2025

• [Docs] Show correct list of vertex ai mistral models

  #12828 merged Jul 21, 2025

• [UI Bug Fix] Show correct guardrails when editing a team

  #12823 merged Jul 21, 2025

• feat: Add Lambda AI provider support

  #12817 merged Jul 21, 2025

• Adding HolmesGPT to projects using LiteLLM

  #12798 merged Jul 21, 2025

• docs(moonshot): correct base url and document CN-specific endpoint

  #12804 merged Jul 21, 2025

• Fix SSO Logout | Create Unified Login Page with SSO and Username/Password Options

  #12703 merged Jul 21, 2025

• docs - vector stores

  #12781 merged Jul 20, 2025

• Litellm fix proxy unit testing p2

  #12779 merged Jul 19, 2025

• UI - support adding links to model hub

  #12776 merged Jul 19, 2025

• Litellm fix proxy unit testing

  #12778 merged Jul 19, 2025

• [LLM Translation] Add Gov Cloud bedrock model pricing and context windows

  #12773 merged Jul 19, 2025

• [LLM Translation] added switchpoint router

  #12777 merged Jul 19, 2025

• fix(proxy): Fix Model Armor project_id initialization order

  #12766 merged Jul 19, 2025

• Allow forwarding clientside headers by model group

  #12753 merged Jul 19, 2025

• [QA] Disable Logging settings for Keys

  #12774 merged Jul 19, 2025

• Fix moonshot/kimi-thinking-preview tool choice support

  #12772 merged Jul 19, 2025

• Feature/track bedrock gov cloud models

  #12771 merged Jul 19, 2025

• [Key Access] Litellm disabled callbacks for UI

  #12769 merged Jul 19, 2025

• UI - Support 'batch' model health checks + make 'team-only' model concept clearer

  #12770 merged Jul 19, 2025

• [JSON Logs] fix ciruclar ref error by adding safe dumps

  #12764 merged Jul 19, 2025

• fix: correct Groq model naming convention for moonshotai/kimi-k2-instruct

  #12768 merged Jul 19, 2025

• feat(proxy_server.py): add model hub to the swagger

  #12767 merged Jul 19, 2025

• [Docs] 1.74.6.rc note

  #12765 merged Jul 19, 2025

• Litellm gemini grounding metadata stream

  #12673 merged Jul 19, 2025

• Bulk Edit Users on UI

  #12763 merged Jul 19, 2025

• [Feat] Backend - Add support for disabling callbacks in request body

  #12762 merged Jul 19, 2025

• Fix Y-axis labels overlap on Spend per Tag

  #12754 merged Jul 19, 2025

• Copy MCP Server name

  #12760 merged Jul 19, 2025

• feat: add v0 provider support

  #12751 merged Jul 19, 2025

• [Feat] UI Vector Stores - Allow adding Vertex RAG Engine, OpenAI, Azure

  #12752 merged Jul 19, 2025

• [Feat] LLM API Endpoint - Expose OpenAI Compatible /vector_stores/{vector_store_id}/search endpoint

  #12749 merged Jul 19, 2025

• [LLM Translation] Added model name formats

  #12745 merged Jul 19, 2025

• [LLM Translation - Redis] fix: redis caching for embedding response models

  #12750 merged Jul 18, 2025

• [LLM Translation] Change System prompts to assistant prompts as a workaround for GH Copilot

  #12742 merged Jul 18, 2025

• build(deps): bump on-headers and compression in /docs/my-website

  #12721 merged Jul 18, 2025

• fix(lowest_latency.py): Handle ZeroDivisionError with zero completion tokens

  #12734 merged Jul 18, 2025

• [Feat] UI - Allow clicking into Vector Stores

  #12741 merged Jul 18, 2025

• Add project_id to cached credentials for VertexAI

  #12661 merged Jul 18, 2025

• feat: integrate Google Cloud Model Armor guardrails

  #12492 merged Jul 18, 2025

• [jais-30b-chat] added model to prices and context window

  #12739 merged Jul 18, 2025

• [Prometheus] Move Prometheus to enterprise folder

  #12659 merged Jul 18, 2025

• Guardrails AI - support llmOutput based guardrails as pre-call hooks

  #12674 merged Jul 18, 2025

• Health check app on separate port

  #12718 merged Jul 18, 2025

• Anthropic - add tool cache control support

  #12668 merged Jul 18, 2025

• /streamGenerateContent - non-gemini model support

  #12647 merged Jul 18, 2025

• Add Hosted VLLM rerank provider integration

  #12738 merged Jul 18, 2025

• Vllm rerank

  #12737 merged Jul 18, 2025

• chore(proxy): loosen rich version from ==13.7.1 to >=13.7.1

  #12704 merged Jul 18, 2025

• [Bug fix] s3 v2 log uploader crashes when using with guardrails

  #12733 merged Jul 18, 2025

• [Bug Fix] QA - Use PG Vector Vector Store with LiteLLM

  #12716 merged Jul 18, 2025

• fixed comment in docs for anthropic provider

  #12725 merged Jul 18, 2025

• [Feat] Add azure_ai/grok-3 model family + Cost tracking

  #12732 merged Jul 18, 2025

• Fix AsyncMock error in team endpoints test

  #12730 merged Jul 18, 2025

• Regenerate Key State Management and Authentication Issues

  #12729 merged Jul 18, 2025

• Teams - allow setting custom key duration + show many user + service account keys have been created

  #12722 merged Jul 18, 2025

• fix(team_endpoints.py): ensure user id correctly added when new team …

  #12719 merged Jul 18, 2025

• feat(internal_user_endpoints.py): new /user/bulk_update endpoint

  #12720 merged Jul 18, 2025

• Litellm encrypt admin UI values - ensures SSO keys work on UI

  #12675 merged Jul 18, 2025

• MCP Gateway Change name for all servers to all available servers for Internal Users

  #12702 merged Jul 18, 2025

• [MCP Gateway] add fix to update object permission on update/delete key/team

  #12701 merged Jul 18, 2025

• [MCP Gateway] added docs for mcp namespacing by URL

  #12700 merged Jul 18, 2025

• [Refactor] Vector Stores - Use class VectorStorePreCallHook for all Vector Store Integrations

  #12715 merged Jul 17, 2025

• [Bug Fix] Always include tool calls in output of trim_messages

  #11517 merged Jul 17, 2025

• [Refactor] Use Existing config structure for bedrock vector stores

  #12672 merged Jul 17, 2025

• [Feat] Proxy - New LLM API Routes /v1/vector_stores and /v1/vector_stores/vs_abc123/search

  #12699 merged Jul 17, 2025

• [Feat] Bedrock Guardrails - Allow disabling exception on 'BLOCKED' action

  #12693 merged Jul 17, 2025

• Fix incorrect environment variable names in Claude code docs

  #12686 merged Jul 17, 2025

• Add Claude Code LiteLLM tutorial

  #12650 merged Jul 17, 2025

• [Liveness/Liveliness probe] add separate health app for liveness probes in files

  #12669 merged Jul 17, 2025

• [Feat] New Vector Store - PG Vector

  #12667 merged Jul 17, 2025

• feat: add input_fidelity parameter for OpenAI image generation

  #12662 merged Jul 16, 2025

• [Bug Fix] SCIM - add GET /ServiceProviderConfig

  #12664 merged Jul 16, 2025

• [Bug Fix] StandardLoggingPayload on cache_hits should track custom llm provider + DD LLM Obs span type

  #12652 merged Jul 16, 2025

• [Feat] UI - Add end_user filter on UI

  #12663 merged Jul 16, 2025

• [Feat] Allow reading custom logger python scripts from s3

  #12623 merged Jul 16, 2025

• [MCP Gateway] Allow MCP sse and http to have namespaced url for better segregation LIT-304

  #12658 merged Jul 16, 2025

• [MCP Gateway] List tools from access list for keys

  #12657 merged Jul 16, 2025

• [MCP Gateway] Allow MCP access groups to be added via the config LIT-312

  #12654 merged Jul 16, 2025

• Fix unused imports in completion_extras transformation

  #12655 merged Jul 16, 2025

• Add GitHub Copilot LiteLLM tutorial

  #12649 merged Jul 16, 2025

• [Bug Fix] grok-4 does not support the stop param

  #12646 merged Jul 16, 2025

• Add groq/moonshotai-kimi-k2-instruct model configuration

  #12648 merged Jul 16, 2025

• [New Model] add together_ai/moonshotai/Kimi-K2-Instruct

  #12645 merged Jul 16, 2025

• Fix bedrock nova micro and lite info

  #12619 merged Jul 16, 2025

• fix: Handle circular references in spend tracking metadata JSON serialization

  #12643 merged Jul 16, 2025

• OpenAI deepresearch models via .completion support

  #12627 merged Jul 16, 2025

• fix(proxy_server.py): fixes for handling team only models on UI

  #12632 merged Jul 16, 2025

• fix(router.py): use more descriptive error message + UI - enable team admins to update member role

  #12629 merged Jul 16, 2025

• [Bug Fix] [Bug]: Knowledge Base Call returning error

  #12628 merged Jul 16, 2025

• (#11794) use upsert for managed object table rather than create to avoid UniqueViolationError

  #11795 merged Jul 16, 2025

• fix: role chaining with webauthentication for aws bedrock

  #12607 merged Jul 16, 2025

• Add input_cost_per_pixel to values in ModelGroupInfo model

  #12604 merged Jul 16, 2025

• Add token pricing for Together.ai Llama-4 and DeepSeek models

  #12622 merged Jul 16, 2025

• Add "keys import" command to CLI

  #12620 merged Jul 16, 2025

• rm claude instant 1 and 1.2 from model_prices_and_context_window.json

  #12631 merged Jul 16, 2025

• [Feat] MCP Gateway - allow using MCPs with all LLM APIs when using /responses with LiteLLM

  #12546 merged Jul 15, 2025

• [Docs] troubleshooting SSO configs

  #12621 merged Jul 15, 2025

• [Bug Fix] Add swagger docs for LiteLLM /chat/completions, /embeddings, /responses

  #12618 merged Jul 15, 2025

• refactor(mcp): Make MCP_TOOL_PREFIX_SEPARATOR configurable from env

  #12603 merged Jul 15, 2025

• add azure blob cache support

  #12587 merged Jul 15, 2025

• Add Copy-on-Click for IDs

  #12615 merged Jul 15, 2025

• [Bug Fix] Include /mcp in list of available routes on proxy

  #12612 merged Jul 15, 2025

• fix(anthropic): fix streaming + response_format + tools bug

  #12463 merged Jul 15, 2025

• feat(gemini): Add custom TTL support for context caching (#9810)

  #12541 merged Jul 15, 2025

• feat: Add envVars and extraEnvVars support to Helm migrations job

  #12591 merged Jul 15, 2025

• refactor(prisma_migration.py): refactor to support use_prisma_migrate - for helm hook

  #12600 merged Jul 15, 2025

• Claude 4 Bedrock /invoke route support + Bedrock application inference profile tool choice support

  #12599 merged Jul 15, 2025

• Control Plane + Data Plane support

  #12601 merged Jul 15, 2025

• [Bug fix] [Bug]: Verbose log is enabled by default

  #12596 merged Jul 15, 2025

• Wildcard model filter

  #12597 merged Jul 15, 2025

• [Feat] Vector Stores - Add Vertex RAG Engine API as a provider

  #12595 merged Jul 15, 2025

• Updated release notes

  #12594 merged Jul 15, 2025

• fix: add implicit caching cost calculation for Gemini 2.x models

  #12585 merged Jul 14, 2025

• [Feat] Add ai21/jamba-1.7 model family pricing

  #12593 merged Jul 14, 2025

• [Feat] New LLM API Integration - Add Moonshot API (Kimi) (#12551)

  #12592 merged Jul 14, 2025

• [Feat] New LLM API Integration - Add Moonshot API (Kimi)

  #12551 merged Jul 14, 2025

• [Feat] Add Moonshot AI / Kimi

  #12589 merged Jul 14, 2025

• Add Bytez to the list of providers in the docs

  #12588 merged Jul 14, 2025

• Add pricing information for Moonshot AI's kimi-k2 model

  #12566 merged Jul 14, 2025

• Litellm release notes 07 12 2025

  #12563 merged Jul 13, 2025

• [MCP Gateway] Ensure we use the same param for specifying groups

  #12561 merged Jul 12, 2025

• UI - v1.74.3-stable QA fixes

  #12559 merged Jul 12, 2025

• fix: handle missing 'env' attribute in MCP server Prisma model

  #12560 merged Jul 12, 2025

• [MCP Gateway] add access group documentation

  #12557 merged Jul 12, 2025

• Fix: Output github copilot verification uri immediately when running in docker.

  #12558 merged Jul 12, 2025

• UI - Model Hub - refactor 'Make Public' to have a select + confirm form

  #12555 merged Jul 12, 2025

• UI - Model Hub Page - minor fixes + improvements (+ Make Model Hub OSS 🚀)

  #12553 merged Jul 12, 2025

• [MCP Gateway] access group fixes on UI for keys and teams

  #12556 merged Jul 12, 2025

• [CI/CD fix] test_redis_caching_multiple_namespaces

  #12552 merged Jul 12, 2025

• Integration: Bytez as a model provider

  #12121 merged Jul 12, 2025

• Fix e2e test

  #12549 merged Jul 12, 2025

• [MCP Gateway] UI headers groups example on connect tab

  #12550 merged Jul 12, 2025

• [MCP Gateway] Allow mcp access groups on test key and tool calls

  #12529 merged Jul 12, 2025

• Prevent writing default user setting updates to yaml (error in non-root env) + Use central team member budget when max_budget_in_team set on UI

  #12533 merged Jul 12, 2025

• [Bug Fix] xai/ translation fix - ensure finish_reason includes tool calls when xai responses with tool calls

  #12545 merged Jul 12, 2025

• chore: Update Vertex AI Model Garden LiteLLM integration tutorial

  #12428 merged Jul 12, 2025

• Align Show Password with Checkbox

  #12538 merged Jul 12, 2025

• Fix e2e test

  #12544 merged Jul 12, 2025

• Consistent layout for Create and Back buttons on all the pages

  #12542 merged Jul 12, 2025

• Team Members - reset budget, if duration set + Prometheus - support tag based metrics

  #12534 merged Jul 12, 2025

• 🐛 Remove deprecated pydantic class Config

  #12528 merged Jul 12, 2025

• docs: Update github.md

  #12509 merged Jul 12, 2025

• [MCP Gateway] Allow using stdio MCPs with LiteLLM

  #12530 merged Jul 12, 2025

• [MCP Gateway] UI Quality check fixes

  #12521 merged Jul 12, 2025

• [MCP Gateway] access group UI object permission fix

  #12523 merged Jul 12, 2025

• [Bug Fix] - QA for MCP Gateway - show the cost config on the root of MCP Settings

  #12526 merged Jul 12, 2025

• [Security Fix] - Dont show pure JWT in "Logs" page on UI

  #12524 merged Jul 11, 2025

• [MCP Gateway] QA - MCP Tool Testing Playground

  #12520 merged Jul 11, 2025

• Litellm mcp groups UI

  #12522 merged Jul 11, 2025

• [Feat] - New guardrail - OpenAI Moderations API

  #12519 merged Jul 11, 2025

• Validation to mcp server name

  #12515 merged Jul 11, 2025

• Add Build and push litellm-non_root to docker-hub-deploy CICD workflow

  #12413 merged Jul 11, 2025

• Litellm mcp access group

  #12514 merged Jul 11, 2025

• [Enterprise] Support tag based mode for guardrails

  #12508 merged Jul 11, 2025

• Litellm mcp access group on UI

  #12470 merged Jul 11, 2025

• Guardrails AI - pre-call + logging only guardrail (pii detection/competitor names) support

  #12506 merged Jul 11, 2025

• Fix tool call handling in Anthropic pass-through adapter

  #12473 merged Jul 11, 2025

• fix bedrock cost calculation for cached tokens

  #12488 merged Jul 11, 2025

79 Pull requests opened by 51 people

• [Feature: Added the support to add api_url in custom_guardrail]

  #12512 opened Jul 11, 2025

• Fix: Handle missing test runner script in LLM translation workflow

  #12516 opened Jul 11, 2025

• [Draft PR] Allow setting disabled callbacks on UI

  #12527 opened Jul 11, 2025

• Improved Provider and Model Name Handling

  #12536 opened Jul 12, 2025

• Added json_schema to optional parameters to include a json schema for structured responses

  #12572 opened Jul 14, 2025

• Show failed token on key deletion

  #12577 opened Jul 14, 2025

• [Bug]: LiteLLM logs certain error messages to stdout instead of stderr

  #12580 opened Jul 14, 2025

• [Docs] Update Hugging Face documentation

  #12582 opened Jul 14, 2025

• Restrict Team settings on Add Model page

  #12617 opened Jul 15, 2025

• [WIP] chore: lazy load the model costs

  #12633 opened Jul 16, 2025

• fix(embeddings): Avoid check for openai if the api base is not the official url

  #12637 opened Jul 16, 2025

• fix: ensure Gemini calls include a user message (fixes #9733)

  #12651 opened Jul 16, 2025

• Make a prompt_id optional for custom_prompt:get_chat_completion_prompt

  #12656 opened Jul 16, 2025

• Convert <think> to reasoning_content

  #12692 opened Jul 17, 2025

• Fix Cohere embedding error when base64 encoding_format is requested

  #12696 opened Jul 17, 2025

• fix(bedrock-converse): cache control not applied to messages with assistant or tool role

  #12697 opened Jul 17, 2025

• ci(lint): black not running with check

  #12698 opened Jul 17, 2025

• fix(model_management_endpoints.clear_cache): only clear database models from cache on model update

  #12709 opened Jul 17, 2025

• Fix internal users table overflow

  #12736 opened Jul 18, 2025

• Add Vertex AI supervised fine-tuning and online prediction integration

  #12758 opened Jul 19, 2025

• Fix RuntimeWarning: coroutine was never awaited in async_client_cleanup

  #12759 opened Jul 19, 2025

• feat: Built-in MCP Server Integration - Server-to-Server MCP with Client Abstraction

  #12790 opened Jul 20, 2025

• #12794 Commit the codes for fixing the team budget max limitation bug.

  #12796 opened Jul 21, 2025

• Fix additional anyOf corner cases for Vertex AI Gemini tool calls - issue #11164

  #12797 opened Jul 21, 2025

• #12800 Commit the codes for fixing the no model_group information in …

  #12801 opened Jul 21, 2025

• Feature/vertex ai fine tuning

  #12835 opened Jul 21, 2025

• [QA] Allow viewing redacted standard callback dynamic params

  #12853 opened Jul 22, 2025

• fix: make gemini and openai responses return reasoning by default

  #12865 opened Jul 22, 2025

• fix: streaming finish_reason returns tool_calls for o1-pro and Cohere models

  #12866 opened Jul 22, 2025

• Remove vector store methods from global scope

  #12885 opened Jul 23, 2025

• feat: Add 'All Org Models' option for API key model access

  #12907 opened Jul 23, 2025

• build(deps-dev): bump form-data from 4.0.1 to 4.0.4 in /tests/proxy_admin_ui_tests/ui_unit_tests

  #12914 opened Jul 23, 2025

• Move Team Selection for a Model to Advance Settings

  #12927 opened Jul 24, 2025

• [QA] Fixes for using Auto Router + UI fixes

  #12983 opened Jul 25, 2025

• honor OLLAMA_API_KEY for ollama_chat

  #12984 opened Jul 25, 2025

• feat: Add Tracing Support For Anthropic/Claude Code With Arize/Phoenix Integration

  #12987 opened Jul 25, 2025

• fix(proxy): fix GCP Model Armor guardrail detection and circular reference issue

  #12991 opened Jul 25, 2025

• Heroku llms

  #12992 opened Jul 25, 2025

• fix: enable metadata parameter for OpenAI API calls

  #12999 opened Jul 25, 2025

• Fix Add Model

  #13010 opened Jul 26, 2025

• Add OpenRouter Qwen models

  #13019 opened Jul 26, 2025

• feat: add streaming tool call filtering

  #13021 opened Jul 27, 2025

• Fix div by zero

  #13025 opened Jul 27, 2025

• fix: Add custom httpx.Client support

  #13054 opened Jul 28, 2025

• Update Railway deploy links and new guide

  #13091 opened Jul 29, 2025

• (Not fully tested, LLM-generated code) fix issue where vertex ai fails to use new credentials after token expiration plus gcloud auth login --update-adc

  #13092 opened Jul 29, 2025

• Add Inference.net provider: docs update, provider config, and test suite

  #13104 opened Jul 29, 2025

• Optionally enable GenAI telemetry following semantic conventions (#1…

  #13105 opened Jul 29, 2025

• change openai api us endpoint

  #13123 opened Jul 30, 2025

• feat(helm): Enhance Helm chart for ArgoCD integration and external resources

  #13125 opened Jul 30, 2025

• Add Vercel AI Gateway provider

  #13144 opened Jul 30, 2025

• Team Member Permissions Page - Access Column Changes

  #13145 opened Jul 30, 2025

• [Bug]: Set user from token user_id for OpenMeter integration

  #13152 opened Jul 30, 2025

• Fix incorrect message index handling on assistant message merging.

  #13153 opened Jul 30, 2025

• Update Pangea Guardrail to support new AIDR endpoint

  #13160 opened Jul 31, 2025

• feat(bedrock): Add S3 file upload and batch processing integration to llm_http_handler

  #13167 opened Jul 31, 2025

• update: remove +1 from comp in rate limiter

  #13176 opened Jul 31, 2025

• Fix: Handle missing 'choices' field in Azure OpenAI responses

  #13201 opened Aug 1, 2025

• Add OAuth2 SSO endpoint for external application integration

  #13227 opened Aug 2, 2025

• Ensure that `function_call_prompt` extends system messages following its current schema

  #13243 opened Aug 3, 2025

• fix: improve Gemini API key masking in debug logs

  #13272 opened Aug 4, 2025

• 🆕 [FEATURE]: Add WandB by Coreweave Inference Endpoints as a hub.

  #13290 opened Aug 5, 2025

• Feat/sambanova embeddings

  #13308 opened Aug 5, 2025

• Fix/gcs cache docs missing for proxy mode

  #13328 opened Aug 6, 2025

• Support Gemini CLI as provider of LiteLLM

  #13331 opened Aug 6, 2025

• [Feat] Add Streaming support for bedrock gpt-oss model family

  #13346 opened Aug 6, 2025

• feat: add parasail provider

  #13349 opened Aug 6, 2025

• Fix unclosed aiohttp client session warnings during concurrent requests

  #13371 opened Aug 7, 2025

• Fix token_counter with special token input

  #13374 opened Aug 7, 2025

• Fix Ollama GPT-OSS streaming with 'thinking' field

  #13375 opened Aug 7, 2025

• Enhance logging for containers to log on files both with usual format and json format

  #13394 opened Aug 7, 2025

• Logging Callback - Langfuse Host param adde

  #13409 opened Aug 8, 2025

• Make sure that no debug information is leaked by router through errors in Cooldown list

  #13413 opened Aug 8, 2025

• Allow unsetting TPM and RPM - Teams Settings

  #13430 opened Aug 8, 2025

• Display Error from Backend on the UI - Keys Page

  #13435 opened Aug 8, 2025

• Fix OCI streaming

  #13437 opened Aug 8, 2025

• feat(usage): aggregated tag daily activity endpoint + UI fallback; MCP optional; tests included

  #13446 opened Aug 8, 2025

• feat: add CometAPI provider support with chat completions and streaming

  #13458 opened Aug 9, 2025

• [LLM Translation] Fix Realtime API endpoint for no intent

  #13476 opened Aug 9, 2025

435 Issues closed by 39 people

• [Bug]: sync with redis for budget / spend happens too infrequently

  #6288 closed Aug 11, 2025

• [Bug]: MonthlyGlobalSpend doesn't exist, Last30dKeysBySpend relation doesn't exist

  #6419 closed Aug 11, 2025

• [Bug]: Does x-ai not support the response_format parameter?

  #6610 closed Aug 11, 2025

• [Bug]: Azure ad token provider does not work

  #6790 closed Aug 11, 2025

• [Bug]: Anthropic usage prompt cache details missing from logging callbacks when streaming

  #7790 closed Aug 11, 2025

• [Bug]: NotFoundError (404) with Custom API Base for Gemini

  #8772 closed Aug 11, 2025

• Self-Hosted DeepSeek Model used in OpenAI-Codex - Endpoint Issue

  #10309 closed Aug 11, 2025

• Critical vulnerabilities identified by PrismaCloud

  #10513 closed Aug 11, 2025

• [Bug]: Somehow optional column is not working in function calling for o4-mini

  #10552 closed Aug 11, 2025

• Getting 'Exception' object has no attribute 'request' in every step of my agent

  #10560 closed Aug 11, 2025

• [Bug]: Small floats in scientific notation get interpreted as strings

  #10564 closed Aug 11, 2025

• [Question] Guardrails not working as expected

  #10565 closed Aug 11, 2025

• [Feature]: `babbage:2023-07-21-v2` in cost tracking

  #10571 closed Aug 11, 2025

• [Bug]: `litellm>=1.65.5` not propagating `model` during streaming

  #10572 closed Aug 11, 2025

• [Bug]: r1 metadata incorrectly marks "supports_function_calling":true,"supports_tool_choice":true

  #10574 closed Aug 11, 2025

• [Bug]: openai does not support parameters: ['reasoning_effort'], for model=gpt-5

  #13402 closed Aug 10, 2025

• [Bug]: Open WebUI auto generated chat titles prevented by Emoji in model name

  #13481 closed Aug 10, 2025

• [Bug]: Bedrock PII Masking issue

  #13180 closed Aug 10, 2025

• [Bug]: Failure in fallback to gpt-4o for gemini requests with image url

  #9816 closed Aug 10, 2025

• [Bug]: gpt-5 - add support for new OpenAI params

  #13391 closed Aug 9, 2025

• [Bug]: JWTs access not working with model groups

  #13453 closed Aug 9, 2025

• [Bug]: MCP Tools are not persisted

  #13461 closed Aug 9, 2025

• [Feature]: Proxy - Allow a button / periodic trigger for reloading pricing data

  #13383 closed Aug 9, 2025

• [Feature]: Support for Chat Completions and Dynamic Guardrails in Swagger

  #13457 closed Aug 9, 2025

• Comet: LiteLLM.Info

  #13467 closed Aug 9, 2025

• [Bug]: `temperature` not dropped in proxy

  #13327 closed Aug 9, 2025

• [Bug]: No module named 'litellm.llms.mistral.chat.transformation'; 'litellm.llms.mistral.chat' is not a package

  #13259 closed Aug 9, 2025

• [Bug]: DBRX - Signature Error with OSS GPT models

  #13304 closed Aug 9, 2025

• [Bug]: Internal Server Error with GET request for /routes litellm 1.74.12 and several other releases

  #13184 closed Aug 9, 2025

• [Bug]: Responses API failed if input containing ResponseReasoningItem

  #13420 closed Aug 9, 2025

• [Bug]: Generic SSO send user to /sso/key/generate

  #13335 closed Aug 9, 2025

• [Bug]: Impossible to choose models in Team

  #13414 closed Aug 9, 2025

• [Feature]: Support `cerebras/gpt-oss` and `openrouter/gpt-oss`

  #13428 closed Aug 9, 2025

• [Feature]: Support for IAM role credentials for scaling Bedrock usage to multiple accounts

  #13450 closed Aug 9, 2025

• [Feature]: Support for GPT-OSS models from litellm for faster inference with Cerebras stack

  #13451 closed Aug 9, 2025

• [Feature]: Support for IAM role credentials for scaling Bedrock usage to multiple accounts

  #13449 closed Aug 9, 2025

• [Feature]: Support for IAM role credentials for scaling Bedrock usage to multiple accounts

  #13448 closed Aug 9, 2025

• [Feature]: Integrating user information generated by LiteLLM with Langfuse

  #7238 closed Aug 9, 2025

• [Bug]: drop_params not working for (image only?) vertex_ai generation

  #9936 closed Aug 9, 2025

• [Feature]: Slack message redaction

  #10233 closed Aug 9, 2025

• [Bug]: async more than 2000 not work

  #10526 closed Aug 9, 2025

• [Bug]: Incorrect GPT-5 Input Token Limit

  #13439 closed Aug 8, 2025

• [Bug]: [since v1.75.2] Error creating standard logging object - can't register atexit after shutdown

  #13424 closed Aug 8, 2025

• [Bug]: After removing a user from a team via the API not able to add them back to that team

  #13422 closed Aug 8, 2025

• [Bug]: Empty list of MCP tools returned on streamable connection

  #13298 closed Aug 8, 2025

• [Bug]: OpenAI gpt-5 responses API caps streaming responses

  #13393 closed Aug 8, 2025

• [Bug]: OpenAI models health check fails

  #13248 closed Aug 8, 2025

• [Bug]: files created without target_model_names are not tracked by batch cost scheduler

  #13364 closed Aug 8, 2025

• [Bug]: 'Prisma' object has no attribute 'litellm_prompttable'

  #13404 closed Aug 8, 2025

• [Bug]: Job "update_spend" raised an exception

  #13204 closed Aug 8, 2025

• [Bug]: Azure gpt-5 response error

  #13398 closed Aug 8, 2025

• [Bug]: Azure OpenAI gpt-5 Tempature Value

  #13397 closed Aug 8, 2025

• [Feature]: Poll ollama for new endpoints

  #979 closed Aug 8, 2025

• [Bug]: Installing litellm[proxy] failed uvloop does not support Windows

  #7731 closed Aug 8, 2025

• [Bug]: Improperly handled exception (misassumed exception has request body)

  #10475 closed Aug 8, 2025

• [Feature > Bedrock/Converse]: Allow manually specifying a model's supported features or enabling all features by default

  #10506 closed Aug 8, 2025

• [Feature]: putting status code in `litellm.APIError.__str__`

  #10511 closed Aug 8, 2025

• [Bug]: OpenAI GPT-5 series does not support "max_tokens" parameter

  #13381 closed Aug 7, 2025

• [Bug]: Migration using non-root image failed

  #13282 closed Aug 7, 2025

• [Bug]: Bedrock Guardrails - Sensitive Data Leaked to Logs When Using SensitiveInformationPolicyConfig

  #12152 closed Aug 7, 2025

• [Bug]: max_tokens parameter causes BedrockException “max_tokens_to_sample: Extra inputs are not permitted” for Claude Opus 4.1 model

  #13301 closed Aug 7, 2025

• [Bug]: Multi-instance rate limiting not working even in version 1.74.9

  #13202 closed Aug 7, 2025

• [Bug]: openai OSS through Groq is not mapped in litellm

  #13351 closed Aug 7, 2025

• [Bug]: Create, Search Vector Store fail

  #13284 closed Aug 7, 2025

• [Bug]: Total tokens do not count cache create input tokens (LiteLLM Proxy)

  #13207 closed Aug 7, 2025

• cleanup - migrate ollama calls to `ollama_chat`

  #5048 closed Aug 7, 2025

• [Bug]: x-litellm-cache-key header not being returned on cache hit

  #8570 closed Aug 7, 2025

• [Bug]: Error when testing the connection

  #10429 closed Aug 7, 2025

• [Bug]: stream handler completion throwing exception with watsonx

  #10457 closed Aug 7, 2025

• [Feature]: Allow specifying deployment id on passthrough

  #10466 closed Aug 7, 2025

• Submitting a file is not working due to

  #10472 closed Aug 7, 2025

• [Bug]: LiteLLM raises 500 when create/update a team member budget

  #10477 closed Aug 7, 2025

• [Feature]: Improved Handling of Langfuse Trace Metadata from Request Headers

  #10480 closed Aug 7, 2025

• [Feature]: Cache tokens for vertex-ai/gemini models!

  #10481 closed Aug 7, 2025

• [Bug]: Editable install fails with pip<21.3

  #10486 closed Aug 7, 2025

• [Bug]: Failed listing MCP tools of a server as a non-admin user

  #13341 closed Aug 6, 2025

• [Bug]: LiteLLM_MCPServerTable.server_name does not exist (missing migration?)

  #13288 closed Aug 6, 2025

• [Bug]: Failed to get available MCP tools - MCP Auth

  #13199 closed Aug 6, 2025

• [Bug]: Tool-Calling of Anthropic via Bedrock not working: Expected 'id' to be a string

  #13124 closed Aug 6, 2025

• [Bug]: tool_calls coming back as null while they should be Array

  #13055 closed Aug 6, 2025

• [Bug]: Listing models should work even when end-user budget is exceeded

  #13286 closed Aug 6, 2025

• [Feature]: Return x-litellm-response-cost header when streaming with include_usage: true

  #12689 closed Aug 6, 2025

• [Bug]: New Databricks Foundation Models databricks-gpt-oss-20b and databricks-gpt-oss-120b failed with error: litellm.APIConnectionError: 'signature'

  #13313 closed Aug 6, 2025

• [Bug]: Mimetype Resolution Error in Bedrock Document Understanding

  #12260 closed Aug 6, 2025

• [Feature]: `function_to_dict` supporting type unions

  #4249 closed Aug 6, 2025

• [Feature]: `function_to_dict` supporting defaulted arguments

  #4250 closed Aug 6, 2025

• [Bug]: inability to use `DeploymentTypedDict` in Pydantic `TypeAdapter` with Python<3.12

  #5664 closed Aug 6, 2025

• [Bug]: Router not respecting TPM limits in concurrent async calls

  #5783 closed Aug 6, 2025

• [Bug]: randomly logs blow up with `router_cooldown_event_callback but _deployment is None for deployment_id=abc123. Doing nothing`

  #5933 closed Aug 6, 2025

• [Feature]: convenience `Enum` for `tool_choice`

  #6091 closed Aug 6, 2025

• [Bug]: `litellm.acompletion` with `stream`ing is unexpectedly manipulating the input `model_list[0]["litellm_params"]`

  #6156 closed Aug 6, 2025

• [Bug]: `litellm.text_completion` not respecting `model_list`

  #6157 closed Aug 6, 2025

• [Feature]: automated nesting when using litellm sdk within langfuse observe() decorated function

  #8423 closed Aug 6, 2025

• Google Authentication Issue

  #8424 closed Aug 6, 2025

• API Keys not displaying after creation

  #8446 closed Aug 6, 2025

• [Feature]: Add Qwen2.5-3B-Instrcut

  #10437 closed Aug 6, 2025

• [Bug]: Azure Audio Transcription & DALL-E (Access denied)

  #10440 closed Aug 6, 2025

• [Bug]: Random problems with docker images using OpenSSL 3.5

  #10444 closed Aug 6, 2025

• [Bug]: gemini request body is not logged

  #10449 closed Aug 6, 2025

• [Feature]: Add "claude-opus-4-1-20250805" in "model_prices_and_context_window.json"

  #13305 closed Aug 5, 2025

• [Bug]: Claude Code count_tokens API is not implemented in the LiteLLM proxy.

  #13252 closed Aug 5, 2025

• [Bug]: LiteLLM Gateway crashes when processing request to Bedrock with tool of JSON Schema that contains circular $ref references

  #13151 closed Aug 5, 2025

• [Bug]: Tool calling broken from expecting "keys" on OpenAITool (v1.74.9)

  #13064 closed Aug 5, 2025

• [Bug]: api request works even if DISABLE_LLM_API_ENDPOINTS=true

  #13095 closed Aug 5, 2025

• [Feature]: (litellm proxy) support /v1/models/{model_id} retrieval

  #13128 closed Aug 5, 2025

• [Bug]: Enforcement of Admin-Only Route Not Working

  #13127 closed Aug 5, 2025

• [Feature]: `litellm --version` not requiring `proxy` extra

  #7975 closed Aug 5, 2025

• [Feature]: Improve Lago integration to better support token based usage billing

  #8243 closed Aug 5, 2025

• [Feature]: adding `ollama/llama3.2` to cost tracking

  #9644 closed Aug 5, 2025

• [Bug]: when new user login to console , there is not have any model to select

  #9678 closed Aug 5, 2025

• litellm.BadRequestError: OpenAIException

  #10083 closed Aug 5, 2025

• [Feature]: Assign Users team role at account creation

  #10109 closed Aug 5, 2025

• [Bug]: Litellm proxy, return empty completion content but rejects to fail

  #10144 closed Aug 5, 2025

• [Bug]: Cost stays zero in the UI and nothing in response headers for ollama custom models

  #10155 closed Aug 5, 2025

• [Bug]: Future attached to a different loop in DualCache.async_batch_get_cache()

  #10376 closed Aug 5, 2025

• [Bug]: dashboard cann't add vllm provider model

  #10400 closed Aug 5, 2025

• Deepseek connection error -- from wren

  #10403 closed Aug 5, 2025

• [Bug]: Gemini response_format list of unions fails with LiteLLM

  #10405 closed Aug 5, 2025

• [Bug]: scheduler keeps creating duplicated pass through route

  #10408 closed Aug 5, 2025

• [Bug]: litellm --health outputs auth_error even the LITELLM_MASTER_KEY envvar is set

  #10412 closed Aug 5, 2025

• Fix "gpt-4o-mini-2024-07-18" entry in "model_prices_and_context_window.json"

  #12913 closed Aug 5, 2025

• [Bug]: Proxy server responses API throws "Invalid content type: <class 'NoneType'>"

  #12670 closed Aug 4, 2025

• [Bug]: Gemini CLI JSON parse error unexpected character "`"

  #12496 closed Aug 4, 2025

• [Bug]: <meta> tag in CLAUDE.md causes 403 when using LiteLLM.

  #12421 closed Aug 4, 2025

• [Bug]: Add service_tier support for /responses API

  #13257 closed Aug 4, 2025

• [Bug]: Versions after 1.73.2 do not automatically route vendors

  #12531 closed Aug 4, 2025

• [Bug]: Missing 'voyage' and 'jina' providers on UI

  #12574 closed Aug 4, 2025

• Organization: is it an enterprise feature?

  #12727 closed Aug 4, 2025

• [Bug]: Deleting a team via UI does not warn or block when related keys exist

  #12947 closed Aug 4, 2025

• [Bug]: Usage.completion_tokens_details.text_tokens isn't serialized

  #13223 closed Aug 4, 2025

• Add "bge-reranker-v2-m3" in "model_prices_and_context_window.json"

  #13121 closed Aug 4, 2025

• [Bug]: MCP server on local docker network is not allowed

  #13132 closed Aug 4, 2025

• [Feature]: Set fallback model for database model in UI

  #13253 closed Aug 4, 2025

• [Bug]: vertex_credentials Not Being Used for Authentication in Vertex AI Calls (Error 400: Invalid API Key)

  #12940 closed Aug 4, 2025

• [Bug]: When using streaming with the OpenAI API, the ID received with each chunk isn't from OpenAI; rather, it's generated by the LiteLLM client.

  #10280 closed Aug 4, 2025

• [Bug]: Unable to Remove Datadog Active Logging Callback Configuration

  #10375 closed Aug 4, 2025

• Issue with LLm lite

  #10377 closed Aug 4, 2025

• [Bug]: How to get Gemini 2.5 cache_read and reasoning_token count while using acompletion with Stream ?

  #10387 closed Aug 4, 2025

• [Bug]: Can't seem to get any Gemini example to work?

  #13241 closed Aug 3, 2025

• [Bug]: *-non_root Images fail on OpenShift due to arbitrary UID/GID0 permissions (needs chgrp 0 && chmod g=u)

  #13208 closed Aug 3, 2025

• [Bug]: ModuleNotFoundError: No module named 'enterprise' When Using litellm==1.48.1 in Google Colab

  #10354 closed Aug 3, 2025

• [Feature]: Support multiple headers inside litellm_key_header_name

  #10355 closed Aug 3, 2025

• [Feature]: Support X-Forwarded-Authorization as well as Authorization

  #10356 closed Aug 3, 2025

• [Bug]: missing enterprise module

  #10361 closed Aug 3, 2025

• [Bug]: ollama_chat not working as expected

  #10362 closed Aug 3, 2025

• OpenAI Rate Limit error on my very first call

  #10365 closed Aug 3, 2025

• [Bug]: Inconsistent behavior with "from google.adk.models.lite_llm import LiteLlm"

  #10369 closed Aug 3, 2025

• [Bug]: Unable to delete Active Logging Callbacks (success_callback issue)

  #10371 closed Aug 3, 2025

• [Bug]: 1.174.12-nightly Auth issue with HTTPS, 305 Errors

  #13211 closed Aug 2, 2025

• [Bug]: NO FALLBACK when streaming and [Bug]: litellm.InternalServerError: AnthropicException - Overloaded. Handle with litellm.InternalServerError.

  #6957 closed Aug 2, 2025

• [Feature]: Add GitHub Copilot as model provider

  #6564 closed Aug 2, 2025

• ratelimit

  #13214 closed Aug 2, 2025

• [Bug]: `extra_headers` are dropped from an embeddings request when provider is `hosted_vllm`/`openai_like`

  #13088 closed Aug 2, 2025

• It seems that this PR changes the behavior of `/key/(un)?block`

  #13136 closed Aug 2, 2025

• [Bug]: Tools streaming broken with ollama_chat

  #6135 closed Aug 2, 2025

• [Feature]: Support Edition methods for Image Generation Models

  #6772 closed Aug 2, 2025

• [Bug]: Gemini tool call result issues

  #13169 closed Aug 1, 2025

• NO FALLBACK when streaming and [Bug]: litellm.InternalServerError: AnthropicException - Overloaded. Handle with `litellm.InternalServerError`.

  #6532 closed Aug 1, 2025

• [Bug]: UI is broken starting from v1.74.9.rc.1

  #13200 closed Aug 1, 2025

• [Bug]: Redis Cluster mode is not support and other issues

  #8159 closed Aug 1, 2025

• [Bug]: litellm.APIError: Error building chunks for logging/streaming usage calculation

  #9983 closed Aug 1, 2025

• [Bug]: OpenRouter Grok Model Exception In Proxy Method

  #10303 closed Aug 1, 2025

• [Feature]: support Streaming STT from fireworks

  #10304 closed Aug 1, 2025

• [Bug]: After sending a streaming request to a custom endpoint, the entire LiteLLM proxy gateway crashed. (If this happens in a production environment, it could cause a major impact.)

  #10305 closed Aug 1, 2025

• [Bug]: incorrect prompt_tokens

  #10311 closed Aug 1, 2025

• [Feature]: support service_tier for o3 and o4-mini

  #10307 closed Aug 1, 2025

• Fix "together-ai-21.1b-41b" entry in "model_prices_and_context_window.json"

  #10315 closed Aug 1, 2025

• [Bug]: Image Model custom cost computation failing

  #10316 closed Aug 1, 2025

• [Bug]: Incorrect cleanup of DB connections when using AWS IAM authentication

  #13120 closed Jul 31, 2025

• [Bug]: Gemini cli not able to access file system to write or create files with LiteLLM

  #12609 closed Jul 31, 2025

• [Bug]: Cache control injection points for Anthropic/Bedrock

  #10226 closed Jul 31, 2025

• [Feature]: Supporting 'thinking'/'reasoning content' output for models from OpenAI

  #13183 closed Jul 31, 2025

• [Bug]: Usage in UI not working for large data

  #13182 closed Jul 31, 2025

• [Bug]: HuggingfaceException - {"error":"Template error: template not found"}

  #12967 closed Jul 31, 2025

• [Bug]: Integration with Langfuse >= 3.0.0 is broken

  #13137 closed Jul 31, 2025

• Grok AI (xAI) support

  #3233 closed Jul 31, 2025

• [Bug]: cannot import name 'get_valid_models' from 'litellm'

  #9755 closed Jul 31, 2025

• [Feature]: Ability to override huggingface endpoint

  #10230 closed Jul 31, 2025

• [Bug]: gemini-2.5-pro-preview-03-25 is incorrectly tagged with supports_reasoning: false

  #10254 closed Jul 31, 2025

• [Bug]: fix caching soft_budget

  #10265 closed Jul 31, 2025

• [Feature]: Model creation should support oauth2 support to be enterprise friendly.

  #10268 closed Jul 31, 2025

• [Feature]: DB Index Selection for Redis Sentinel

  #10276 closed Jul 31, 2025

• [Feature]: Support Passing Lakera Project ID

  #10277 closed Jul 31, 2025

• [Feature]: Apply Guardrails to specific models

  #10279 closed Jul 31, 2025

• [Feature]: Support for gpt-4o-mini-tts parameters and cost tracking in Azure

  #10285 closed Jul 31, 2025

• "litellm.APIConnectionError: 'name' & "litellm.APIConnectionError: OllamaException

  #10291 closed Jul 31, 2025

• [Bug]: Error: Too many pages of data (>10). Please select a smaller date range.

  #13083 closed Jul 30, 2025

• [Feature]: Add mcp server health check

  #13068 closed Jul 30, 2025

• [Bug]: API model not available error while generating data

  #13041 closed Jul 30, 2025

• [Bug]: Entire database wipe when deploying proxy

  #13046 closed Jul 30, 2025

• [Bug]: /v1 not working for model discovery

  #13078 closed Jul 30, 2025

• [Bug]: Bedrock Still generate tool call even when no tools are provided

  #13053 closed Jul 30, 2025

• [Bug]: prisma_client is none

  #10032 closed Jul 30, 2025

• [Bug]: UI Bug: Model Filter dropdown appears behind model list on Models page

  #10217 closed Jul 30, 2025

• [Feature]: could you add the llm into project

  #10219 closed Jul 30, 2025

• Question: Is it possible to change location to europe for Google Gemini provider?

  #10220 closed Jul 30, 2025

• [Bug]: Keys Table UI Issues After Creating a Key

  #10222 closed Jul 30, 2025

• [Feature]: Configuration syntax checker

  #10223 closed Jul 30, 2025

• [Feature]:Integration of Local Ollama Embedding With LiteLLM

  #10225 closed Jul 30, 2025

• [Bug]: New Gemini Models Have "0" as the token costs

  #10242 closed Jul 30, 2025

• [Bug]: openrouter llama-4-scout

  #10246 closed Jul 30, 2025

• No error message, but error?

  #10248 closed Jul 30, 2025

• [Bug]: Global usage stopped working

  #13086 closed Jul 29, 2025

• [Feature]: Support for Computer tool use for via bedrock APIs.

  #12884 closed Jul 29, 2025

• [Bug]: token counter does not expect prefix

  #11791 closed Jul 29, 2025

• [Feature]: Add MCP Client Header version

  #11966 closed Jul 29, 2025

• MCP dependency version is fixed?

  #13080 closed Jul 29, 2025

• Add "openrouter/grok-4" in "model_prices_and_context_window.json"

  #13017 closed Jul 29, 2025

• [Bug]: The Gemini Custom API request has an incorrect authorization format

  #13074 closed Jul 29, 2025

• [Bug]: Claude Code (via Bedrock): BadRequestError in cost_calculator.py — model is missing during passthrough logging

  #12933 closed Jul 29, 2025

• [Bug]: Error when attempting to use /chat/completions when using Azure's `v1-preview` API Version

  #12945 closed Jul 29, 2025

• Unable to load gemini 2.0 flash in litellm

  #13075 closed Jul 29, 2025

• Handling `mycustom/*` by `{provider}/*`

  #10357 closed Jul 29, 2025

• [Feature]: Support AI Studio Imagen4 models

  #13058 closed Jul 29, 2025

• [Bug]: The model gemini-2.5-flash with the merge_reasoning_content_in_choices parameter does not work in version v1.72.2-stable

  #11699 closed Jul 29, 2025

• [Bug]: No scrollbar in the "MCP Tools" if there are more tools from the MCP servers

  #12785 closed Jul 29, 2025

• [Feature]: Ollama schema structured output

  #7131 closed Jul 29, 2025

• [Bug]: Presidio integration failing and making inference return 500

  #10137 closed Jul 29, 2025

• [Feature]: Compatible with fastapi-users

  #10200 closed Jul 29, 2025

• Reduce Premium Copilot requests

  #12859 closed Jul 28, 2025

• [Feature]: add support for vllm rerank by litellm server proxy

  #13044 closed Jul 28, 2025

• [Bug]: Litellm times out after a couple of tasks

  #10180 closed Jul 28, 2025

• Error in LLM response

  #10182 closed Jul 28, 2025

• Integration with watsonX failing

  #10183 closed Jul 28, 2025

• litellm API connection issue

  #10184 closed Jul 28, 2025

• [Bug]: Can't Connect to Postgres Database with Client Certificate and Key

  #10187 closed Jul 28, 2025

• Google-adk with Tavily Search litellm.exceptions.BadRequestError: litellm.BadRequestError: OpenAIException - length of prompt_tokens (141644) must be less than max_seq_len (32768).

  #10188 closed Jul 28, 2025

• [Bug]: Spurious `role: assistant` for Anthropic Vertex tool call streaming in OpenAI Chat Completions format

  #12616 closed Jul 27, 2025

• [Feature]: MCP add support for multiple custom headers

  #12895 closed Jul 26, 2025

• [Bug]: Able to get the response headers from OpenAI but returns None for GCP

  #12986 closed Jul 26, 2025

• [Bug]: cannot pickle 'coroutine' object (again in v1.74.7)

  #12921 closed Jul 26, 2025

• [Bug]: Error committing spend updates

  #13000 closed Jul 26, 2025

• [Feature]: Add design policy document for huge utils.py maintenance

  #3345 closed Jul 26, 2025

• [Bug]: gemini 2.0 Flash GA is erroring due to context caching not being supported

  #8296 closed Jul 26, 2025

• [Feature]: Allow creating dynamic url's as passthrough endpoints

  #9950 closed Jul 26, 2025

• [Bug]: default_team_disabled setting is not work

  #10151 closed Jul 26, 2025

• [Bug]: 401 Authentication Error with custom_llm_provider: openai, completion_url, and api_key for Custom Endpoint

  #10154 closed Jul 26, 2025

• Ollama Remoto: Chamada /api/show ignora OLLAMA_HOST, exigindo OLLAMA_API_BASE redundante

  #10158 closed Jul 26, 2025

• 北京

  #10159 closed Jul 26, 2025

• [Feature]: Anthropic native Responses API Support

  #10160 closed Jul 26, 2025

• [Bug]: CVE SQL Injection

  #12402 closed Jul 25, 2025

• [Bug]: Users not deleted from team

  #12971 closed Jul 25, 2025

• [Bug]: modifying `retry_policy` in config.yaml does not reflect changes

  #12855 closed Jul 25, 2025

• [Bug]: User Budget Not Enforced for Virtual Keys Belonging to Teams

  #12905 closed Jul 25, 2025

• [Feature]: add gemini 2.5 GA models

  #11801 closed Jul 25, 2025

• [Bug]: AzureException BadRequestError – Invalid 'messages[2].tool_calls[0].id' (length 41) when switching from Gemini→Azure

  #11392 closed Jul 25, 2025

• [Bug]: ensure cached_token_details is always an int - if set

  #12924 closed Jul 25, 2025

• [Bug]: chat completions api gives 500 when llm provider is aiohttp_openai

  #8708 closed Jul 25, 2025

• [Bug]: litellm /v1/files error

  #10117 closed Jul 25, 2025

• [Bug]: Caching OpenAI (and potentially other) clients is causing memory-leaks when using Router and multiple threads.

  #10126 closed Jul 25, 2025

• Fix "ollama/llama2:70b" entry in "model_prices_and_context_window.json"

  #10128 closed Jul 25, 2025

• [Bug]: HTTPX dependency outdated

  #10133 closed Jul 25, 2025

• [Bug]: LiteLLM Router carries over completion parameters across requests

  #10136 closed Jul 25, 2025

• [Bug]: Evaluate Multiple LLM Providers with LiteLLM cookbook

  #10143 closed Jul 25, 2025

• [Bug]: ProxyException Unknown Error Occurred

  #10145 closed Jul 25, 2025

• LiteLLM Trademark violators/squatters?

  #10146 closed Jul 25, 2025

• [Feature]: allow for DATABASE_URL being read from a secret instead of username/password/etc.

  #10147 closed Jul 25, 2025

• issue

  #12954 closed Jul 24, 2025

• [Bug]: Vertex AI image generation ignores aspect_ratio parameter

  #12690 closed Jul 24, 2025

• [Bug]: Gemini API 404 Error Due to Incorrect URL Generation

  #12427 closed Jul 24, 2025

• [Bug]: Docker buildx manifest attestations produces `unknown/unknown` image arch in GHCR

  #12412 closed Jul 24, 2025

• [Bug]: SSL CERTIFICATE_VERIFY_FAILED issue with Sentry integration

  #11180 closed Jul 24, 2025

• [Bug] User Header from OpenWebUI Not Appearing in Spend Logs

  #12893 closed Jul 24, 2025

• Request to release a tagged helm chart version

  #10135 closed Jul 24, 2025

• Will there be a leak issue if the stream output is terminated early?

  #12861 closed Jul 24, 2025

• [Feature]: Support assuming an AWS IAM Role for S3 Bucket logging

  #10278 closed Jul 24, 2025

• [Bug]: AttributeError: 'Choices' object has no attribute 'logprobs'

  #12860 closed Jul 24, 2025

• Add "llama3.2:1b" in "model_prices_and_context_window.json"

  #6170 closed Jul 24, 2025

• [Bug]: Windows Compatibility Issue with uvloop

  #7783 closed Jul 24, 2025

• [Bug]: An empty streaming response from GROQ without any error raised.

  #9296 closed Jul 24, 2025

• [Bug]:

  #9511 closed Jul 24, 2025

• [Bug]: Error when using custom model_name in spend/calculate

  #10050 closed Jul 24, 2025

• [Feature]: Support for AzureAI Image Generation Models

  #10057 closed Jul 24, 2025

• [Bug]: o4-mini and o3 don't support stop_sequences parameter

  #10080 closed Jul 24, 2025

• [Feature]: Please support gemini-2.5-pro-exp-03-25 on Google's AI Studio

  #10074 closed Jul 24, 2025

• [Feature]: Requests Per Day (rpd) proxy setting (useful for Google models)

  #10090 closed Jul 24, 2025

• [Feature]: Flex Processing cost tracking

  #10091 closed Jul 24, 2025

• [Feature]: Skip credential check using vertex_ai

  #10092 closed Jul 24, 2025

• [Bug]: UX: Admin GUI: Page scrolling / mouse behavior can cause settings corruption

  #10097 closed Jul 24, 2025

• [Bug]: Missing model mapping for eu.anthropic.claude-3-7-sonnet-20250219-v1:0

  #10100 closed Jul 24, 2025

• [Feature]: Automatically prepend bedrock region

  #10105 closed Jul 24, 2025

• [Bug]: Anthropic models never return finish_reason='stop' when using acompletion with response_format=<Pydantic Model> and stream=True

  #10103 closed Jul 24, 2025

• [Feature]: Add dark mode to UI

  #10110 closed Jul 24, 2025

• litellm auto-downloads files from github

  #12896 closed Jul 23, 2025

• [Bug]: Can't edit Default user settings via the UI (500 internal server error, read-only filesystem)

  #12857 closed Jul 23, 2025

• [Bug]: rm claude 2, 2.1, claude-3-sonnet-20240229 from model_prices_and_context_window.json

  #12863 closed Jul 23, 2025

• Display user defined model name in usage report

  #12887 closed Jul 23, 2025

• Failed to use deepseek

  #7646 closed Jul 23, 2025

• [Feature]: How to batch configure multiple openrouter models in config.yaml for litellm proxy server?

  #8358 closed Jul 23, 2025

• [Feature]: Allow model-level api_key and api_base config for Databricks and Snowflake (multi-account support)

  #9580 closed Jul 23, 2025

• [Bug]: Gemini tool call "default" errors

  #9793 closed Jul 23, 2025

• [Bug]: supports_vision function returns false for gemini/gemini-2.5-pro-exp-03-25

  #9842 closed Jul 23, 2025

• [Feature]: Llama 4 Scout and Maverick API Service vertex_ai

  #10016 closed Jul 23, 2025

• [Bug]: Organization admin can't do much

  #10036 closed Jul 23, 2025

• LiteLLM.Info: If you need to debug this error, use `litellm._turn_on_debug()'.

  #10041 closed Jul 23, 2025

• test

  #10046 closed Jul 23, 2025

• 【Feature Request】Smart Model Routing Based on Query Content

  #10049 closed Jul 23, 2025

• [Bug]: Default Credentials and Internal User Admin Rights differences

  #10078 closed Jul 23, 2025

• [Feature]: Support Computer Control from Anthropic claude-3-5-sonnet-20240620

  #6391 closed Jul 22, 2025

• [Feature]: Recraft API - Add support for Image Edits endpoint

  #12834 closed Jul 22, 2025

• [Feature]: Infra - DB Migration should use `prisma migrate`

  #12104 closed Jul 22, 2025

• [Bug]: Azure KeyVault not in images

  #12846 closed Jul 22, 2025

• [Bug]: Multiple HIGH/CRITICAL CVEs in the LiteLLM non-root image

  #12165 closed Jul 22, 2025

• 严格按照自动化部署的视频中的操作，但视频末尾在实际操作时redis报错

  #12444 closed Jul 22, 2025

• [Feature]: Add support for Hyperbolic as an LLM provider

  #12014 closed Jul 22, 2025

• [Bug]: PII Masking with bedrock

  #10971 closed Jul 22, 2025

• [Feature]: Add cache_control parameter filtering for OpenRouter provider to prevent 404 errors

  #12787 closed Jul 22, 2025

• [Bug]: `litellm.acompletion` retrying does not work

  #12830 closed Jul 22, 2025

• [Bug]: Virtual Key Budget Limit Not Blocking AWS Bedrock Pass-through Requests

  #12789 closed Jul 22, 2025

• [Bug]: Calling /add/allowed_ip when there is no database causes internal server error and locks out everyone but the provided IP

  #12815 closed Jul 22, 2025

• [Bug]: Sonnet 3.5 incorrectly reporting support of `reasoning_effort`

  #12833 closed Jul 22, 2025

• [Feature]: End User Tracking with Vertex AI.

  #12799 closed Jul 22, 2025

• [Bug]: IBM Watsonx - invalid tool choice parameter name #9979

  #12803 closed Jul 22, 2025

• [Bug]: fallbacks on another provider are not working

  #12820 closed Jul 22, 2025

• I got this message when i start the ditto

  #12844 closed Jul 22, 2025

• [Feature]: add library stubs to LiteLLM

  #1540 closed Jul 22, 2025

• [Bug]: When restarting my docker container, no more data or Models

  #6634 closed Jul 22, 2025

• [Bug]: connects to langfuse even it is not configured

  #7732 closed Jul 22, 2025

• [Feature]: Support Embeddings in Infinity

  #8764 closed Jul 22, 2025

• [Bug]: anthropic-beta header not being forwarded on litellm-proxy

  #9016 closed Jul 22, 2025

• [Bug]: [DOCKER] OCI runtime exec failed: exec failed: unable to start container process: exec: "curl": executable file not found in $PATH: unknown

  #9479 closed Jul 22, 2025

• [Bug]: Bedrock guardrails analysing (and blocking) old+system messages

  #9882 closed Jul 22, 2025

• [Feature]: integrate with default opentelemetry exporter configured by `opentelemetry-instrument`

  #9901 closed Jul 22, 2025

• [Feature]: config infinity rerank on UI

  #10007 closed Jul 22, 2025

• [Bug]: Get error 400 when testing the multiagentworkflow example code from the Docs,

  #10015 closed Jul 22, 2025

• [Feature]: respect http_proxy, https_proxy env variables

  #10014 closed Jul 22, 2025

• [Bug]: Databricks claims supports_vision=False but actually supports vision

  #10017 closed Jul 22, 2025

• Add "qwen-vl-plus" in "model_prices_and_context_window.json"

  #12398 closed Jul 21, 2025

• [Feature]: ReCraft api support

  #12822 closed Jul 21, 2025

• [Feature]: Group MCP servers into namespaces and serve in different endpoints

  #12374 closed Jul 21, 2025

• [Bug]: groq/qwen/qwen-qwq-32b is deprecated by groq

  #12825 closed Jul 21, 2025

• [Bug]: API Endpoint for GitHub Copilot should not be hard-coded

  #12726 closed Jul 21, 2025

• [Bug]: All mcp servers are getting listed for key generated to a specific team

  #12568 closed Jul 21, 2025

• [Query] litellm-proxy support for "Mistral Small 3.1" model via Vertex AI provider

  #12535 closed Jul 21, 2025

• [Bug]: no dropdown list for guardrails when editing a team

  #12786 closed Jul 21, 2025

• [Bug]: Unable to route from Anthropic's `/v1/messages` to custom handler

  #12816 closed Jul 21, 2025

• Inconsistent custom_llm_provider in _hidden_params for vertex_ai models

  #10181 closed Jul 21, 2025

• [Bug]: LiteLLM call failed: litellm.APIError: APIError: OpenAIException - 'str' object has no attribute 'model_dump'

  #7730 closed Jul 21, 2025

• [Feature]: Allow extending `litellm.custom_provider_map` using `entrypoints`

  #7733 closed Jul 21, 2025

• [Bug]: DailyTagSpend view missing error

  #8117 closed Jul 21, 2025

• [Bug]: If a request is received where max_tokens is less than thinking budget

  #9001 closed Jul 21, 2025

• [Feature]: Batch API Functionality for OpenAI, Anthropic

  #9680 closed Jul 21, 2025

• [Bug]: no github model is supported

  #9740 closed Jul 21, 2025

• [Bug]: Unable to update public model name of Azure models.

  #9973 closed Jul 21, 2025

• [Feature]: Work with cloudflare gateway on google-ai-studio API

  #9975 closed Jul 21, 2025

• [Bug]: 'Exception' object has no attribute 'request'

  #9977 closed Jul 21, 2025

• [Bug]: IBM Watsonx - invalid tool choice parameter name

  #9979 closed Jul 21, 2025

• [Bug]: Incompatibility with Open WebUI (model list shows embedding models; hard-coded owned_by)

  #9982 closed Jul 21, 2025

• [Bug]: Stream + tools broken with Ollama

  #7094 closed Jul 20, 2025

• litellm.exceptions.BadRequestError

  #9944 closed Jul 20, 2025

• [Bug]: litellm.exceptions.APIConnectionError: litellm.APIConnectionError: Ollama_chatException - {"error":"json: cannot unmarshal array into Go struct field ChatRequest.messages.content of type string"}

  #9959 closed Jul 20, 2025

• Add `openrouter/switchpoint/router` in "model_prices_and_context_window.json"

  #12613 closed Jul 19, 2025

• [Bug]: Google Cloud Model Armor error

  #12757 closed Jul 19, 2025

• [Bug]: Azure gpt-4.1 doesn't support json_schema response format

  #12705 closed Jul 19, 2025

• [Bug]: No grounding metadata on gemini using stream:true

  #10237 closed Jul 19, 2025

• [Bug]: Response format support for Azure models (> 4.1)

  #12708 closed Jul 19, 2025

• [Bug]: Error when calling Gemini model

  #9519 closed Jul 19, 2025

• Potential bugs in python Optimization mode: Use of assert

  #9879 closed Jul 19, 2025

• [Bug]: Can't edit fallbacks on the UI. Changes not saved.

  #9899 closed Jul 19, 2025

• [Possible causes of bugs] Mutable object passed to constructor arguments is assigned to instance variables

  #9928 closed Jul 19, 2025

• [Feature]: support 'auth: true' on passthrough api endpoint

  #9951 closed Jul 19, 2025

• [Bug]: Sending system messages to GitHub Copilot causes Internal Server Errors

  #12724 closed Jul 18, 2025

• [Bug]: ZeroDivisionError in lowest_latency.py: float division by zero

  #12641 closed Jul 18, 2025

• [Bug]: GCP Service Account Credentials not refreshing appropriately when set via env var

  #9863 closed Jul 18, 2025

• LLM issue while using gpt-4o-mini

  #12723 closed Jul 18, 2025

• Add "azure_ai/jais-30b-chat" in "model_prices_and_context_window.json"

  #12712 closed Jul 18, 2025

• [Bug]: Usage issues w/ gemini-cli + VLLM

  #12562 closed Jul 18, 2025

• [Issue]: The new `rich` dependency hard pin to 13.7.1 makes it hard to work with other libraries

  #10673 closed Jul 18, 2025

• [Bug]: S3 success-log uploader crashes with “Circular reference detected”

  #12731 closed Jul 18, 2025

• Add "azure_ai/grok-3" in "model_prices_and_context_window.json"

  #12713 closed Jul 18, 2025

• Add "azure_ai/grok-3-mini" in "model_prices_and_context_window.json"

  #12714 closed Jul 18, 2025

• [Bug]: vertex ai deepseek r1 does not work with proxy

  #12717 closed Jul 18, 2025

• [Bug]: `DeprecationWarning: Use 'content=<...>' to upload raw bytes/text content`

  #5986 closed Jul 18, 2025

• [Bug]: Inconsistent response_format handling between Fireworks AI models

  #7533 closed Jul 18, 2025

• [Bug]: Client Hangs as Transcription errors are not passed to client

  #9712 closed Jul 18, 2025

• [Bug]: /key/info fails on litellm master key

  #9861 closed Jul 18, 2025

• [Bug]: `sagemaker_chat` provider does not correctly pass the `model_id` for the SageMaker Inference Component

  #9909 closed Jul 18, 2025

• [Feature]: Context Caching for Vertex AI

  #6898 closed Jul 17, 2025

• [Bug]: Took too long to import litellm

  #6175 closed Jul 17, 2025

• [Bug]: #7594 broke typing on `Router.acompletion`

  #7641 closed Jul 17, 2025

• [Feature]: Add supports_response_schema for deepseek/deepseek-chat

  #7951 closed Jul 17, 2025

• [Bug]: bedrock guardrails always on

  #8239 closed Jul 17, 2025

• [Bug]: turn_off_message_logging sometimes not redacting output

  #9507 closed Jul 17, 2025

• [Bug]: S3 Config Fails When Using Proxy Custom Hooks

  #12614 closed Jul 16, 2025

• [Feature]: set `modify_params` via environments and drop empty messages

  #9946 closed Jul 16, 2025

• [Bug]: Custom Adapter in passthrough does not work

  #12653 closed Jul 16, 2025

• [Bug]: grok-4 in litellm proxy raises: Argument not supported on this model: stop

  #12635 closed Jul 16, 2025

• [Feature]: Tool Calling support of kimi-k2 via together_ai

  #12639 closed Jul 16, 2025

• not working

  #12638 closed Jul 16, 2025

• [Bug]: circular reference

  #12634 closed Jul 16, 2025

• [Feature]: Add support for OpenAI Deep Research in Bridge for /chat/completion → /responses API

  #12105 closed Jul 16, 2025

• [Bug]: Bedrock inference profiles not mapped for token counter

  #12269 closed Jul 16, 2025

• [Bug]: Knowledge Base Call returning error

  #11404 closed Jul 16, 2025

• [Bug]: remove claude instant 1 & 1.2 from model_prices_and_context_window.json

  #12630 closed Jul 16, 2025

• [Bug]: Cross account webauthn not working

  #12583 closed Jul 16, 2025

• [Bug]: LiteLLM fails to keep model provider

  #9495 closed Jul 16, 2025

• [Bug]: Failures (and preceding prompts) not reported in langfuse

  #9846 closed Jul 16, 2025

• pr_agent.algo.ai_handlers.litellm_ai_handler:chat_completion:319 - Error during LLM inference: litellm

  #9849 closed Jul 16, 2025

• [Feature]: UI - Allow filtering keys by key alias

  #9866 closed Jul 16, 2025

• [Bug]: /mcp missing from the output of /routes

  #12608 closed Jul 15, 2025

• [Feature]: Custom TTL for Gemini Context Caching

  #9810 closed Jul 15, 2025

• [Bug]: Structured Outputs on Bedrock Haiku 3.5 breaks often !!

  #11751 closed Jul 15, 2025

• [Bug]: Bedrock /invoke route doesn't work with claude 4

  #12366 closed Jul 15, 2025

• [Bug]: enable_json_schema_validation not documented on config page

  #12518 closed Jul 15, 2025

• [Bug]: Verbose log is enabled by default

  #12539 closed Jul 15, 2025

• [Bug]: aws_bearer_token not work in proxy model

  #12537 closed Jul 15, 2025

• [Bug]: dashscope/qwen: LLM Provider NOT provided

  #12505 closed Jul 15, 2025

• OpenAI has implemented a new parsing call `client.beta.chat.completions.parse` which supplants/augments the previous (and somewhat standardized chat_completion call). When passing in a pydantic object to response_format it returns an actual pydantic object in the output (not just the json which can be loaded into the pydantic object)

  #5223 closed Jul 15, 2025

• [Bug]: can't use huggingface embeddings via locally hosted `text-embeddings-inference`, fails on getting `model_info`

  #6711 closed Jul 15, 2025

• [Feature]: Add API v2 support for Cohere

  #6980 closed Jul 15, 2025

• [Bug]: calling completion with fallbacks multiple times cause `RuntimeError: Event loop is closed`

  #9133 closed Jul 15, 2025

• [Feature]: Databricks claude-3-7-sonnet Support passing thinking blocks back to claude

  #9790 closed Jul 15, 2025

• [Bug]: `get_litellm_model_info` looks for `base_model` under the wrong key for Azure deployed models

  #9818 closed Jul 15, 2025

• [Bug]: Deepinfra setup via UI

  #9819 closed Jul 15, 2025

• [Feature]: VoyageAI reranking

  #9823 closed Jul 15, 2025

• [Bug]: Wrong Gemini 2.5 cost calculation

  #11156 closed Jul 14, 2025

• [Bug]: API key error for new AI21 Jamba 1.7 models

  #12517 closed Jul 14, 2025

• [Feature]: Add Moonshot AI platform support.

  #12547 closed Jul 14, 2025

• 12

  #12565 closed Jul 14, 2025

• test

  #12567 closed Jul 14, 2025

• [Feature]: Error: This model isn't mapped yet. model=us.deepseek.r1-v1:0

  #9126 closed Jul 14, 2025

• [Bug]: Error in completion_cost when checking the cost of a openai model response with aws_region_name set

  #9803 closed Jul 14, 2025

• [Bug]: json_schema gets converted to tool calling from azure 2025-01-01-preview versions

  #9806 closed Jul 14, 2025

• [Feature]: Add support for assistants API cost tracking on /azure passthrough routes

  #9613 closed Jul 13, 2025

• [Bug]: More efficient healthcheck prompt

  #9788 closed Jul 13, 2025

• [Bug]: adding models through web UI with together.ai provider fails with "Unmapped LLM provider for this endpoint" error

  #9789 closed Jul 13, 2025

• [Bug]: incorrect `tool_result` block transform in Anthropic unified endpoint

  #12404 closed Jul 12, 2025

• [Bug]: `JSONDecodeError` from Anthropic unified endpoint

  #12403 closed Jul 12, 2025

• [Bug]: Pydantic Deprecation Warnings

  #12025 closed Jul 12, 2025

• [Bug]: Support bedrock cost tracking with unknown region

  #9306 closed Jul 12, 2025

• [Bug]: "litellm.BadRequestError: LLM Provider NOT provided" despite openai/ prefix

  #9408 closed Jul 12, 2025

• [Bug]: Docker error with litellm:main-v1.65.4-nightly

  #9777 closed Jul 12, 2025

• Add boolean in API to turn off logs

  #12510 closed Jul 11, 2025

• [Bug]: Wrong cost calculation for caching with Sonnet 4 from Bedrock

  #12258 closed Jul 11, 2025

• [Bug]: Tracked AWS Bedrock spend seems incorrect

  #12313 closed Jul 11, 2025