v1.73.6.rc.1

What's Changed

• [⚡️ Python SDK Import] - 2 second faster import times by @ishaan-jaff in #12135
• 🧹 Refactor init.py to use a model registry by @ishaan-jaff in #12138
• Revert "🧹 Refactor init.py to use a model registry" by @ishaan-jaff in #12141
• [⚡️ Python SDK import] - reduce python sdk import time by .3s by @ishaan-jaff in #12140
• /v1/messages - Remove hardcoded model name on streaming + Tags - enable setting custom header tags by @krrishdholakia in #12131
• UI QA Fixes - prevent team model reset on model add + return team-only models on /v2/model/info + render team member budget correctly by @krrishdholakia in #12144

Full Changelog: v1.73.6-nightly...v1.73.6.rc.1

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.6.rc.1

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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	190.0	207.64485716560858	6.320692412920188	0.0	1890	0	167.53384199989796	1605.5699650000292
Aggregated	Passed ✅	190.0	207.64485716560858	6.320692412920188	0.0	1890	0	167.53384199989796	1605.5699650000292

v1.73.6-nightly

Full Changelog: v1.73.6.rc-draft...v1.73.6-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.6-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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	190.0	210.86720644249243	6.277304583930227	0.0	1878	0	166.350910999995	6235.332546000023
Aggregated	Passed ✅	190.0	210.86720644249243	6.277304583930227	0.0	1878	0	166.350910999995	6235.332546000023

What's Changed

• fix(proxy): Fix test_mock_create_audio_file by adding managed_files hook by @colesmcintosh in #12072
• Enhance CircleCI integration in LLM translation testing workflow by @colesmcintosh in #12041
• Inkeep searchbar and chat added to the Docs by @NANDINI-star in #12030
• [Fix] Redis - Add better debugging to see what variables are set by @ishaan-jaff in #12073
• Fix today selector date mutation bug in dashboard components by @colesmcintosh in #12042
• Responses API - Add reasoning content support for non-OpenAI providers by @ryan-castner in #12055
• Raise clearer error on anthropic unified route + add new new_key param for regenerating key by @krrishdholakia in #12087
• Refactor: bedrock passthrough fixes - migrate to Passthrough SDK by @krrishdholakia in #12089
• Fix Azure-OpenAI Vision API Compliance by @davis-featherstone in #12075
• [Bug Fix] Bedrock Guardrails - Ensure PII Masking is applied on response streaming or non streaming content when using post call by @ishaan-jaff in #12086
• fix(docs): Remove unused dotenv dependency from docusaurus config by @colesmcintosh in #12102
• [Fix] MCP - Ensure internal users can access /mcp and /mcp/ routes by @ishaan-jaff in #12106
• fix: handle provider_config type error in passthrough error handler by @colesmcintosh in #12101
• Add o3 and o4-mini deep research models by @krrishdholakia in #12109
• [Bug Fix] Anthropic - Token Usage Null Handling in calculate_usage by @Gum-Joe in #12068
• fix: change cost calculation logs from INFO to DEBUG level by @colesmcintosh in #12112
• fix: set logger levels based on LITELLM_LOG environment variable by @colesmcintosh in #12111
• [Feat] Add Bridge from generateContent <> /chat/completions by @ishaan-jaff in #12081
• [Docs] - Show how to use fallbacks with audio transcriptions endpoints by @ishaan-jaff in #12115
• [Bug Fix] Fix handling str, bool types formock_testing_fallbacks on router using /audio endpoints by @ishaan-jaff in #12117
• Adding Feature: Palo Alto Networks Prisma AIRS Guardrail by @jroberts2600 in #12116
• [Bug Fix] Exception mapping for context window exceeded - should catch anthropic exceptions by @ishaan-jaff in #12113
• docs(GEMINI.md): add development guidelines and architecture overview by @colesmcintosh in #12035
• [Bug fix] Router - handle cooldown_time = 0 for deployments by @ishaan-jaff in #12108
• [Feat] Add Eleven Labs - Speech To Text Support on LiteLLM by @ishaan-jaff in #12119
• Revert "fix: set logger levels based on LITELLM_LOG environment variable" by @ishaan-jaff in #12122
• Fix Braintrust integration: Adds model to metadata to calculate cost and corrects docs by @ohmeow in #12022
• [Fix] Change Message init type annotation to support other roles by @amarrella in #11942
• Add "Get Code" Feature by @NANDINI-star in #11629
• Bedrock Passthrough cost tracking (/invoke + /converse routes - streaming + non-streaming) by @krrishdholakia in #12123
• feat: add local LLM translation testing with artifact generation by @colesmcintosh in #12120
• [Feat] introduce new environment variable NO_REDOC to opt-out Redoc by @zhangyoufu in #12092
• Fix user-team association issues in LiteLLM proxy by @colesmcintosh in #12082
• feat: enhance redaction functionality for EmbeddingResponse by @bougou in #12088
• De-duplicate models in team settings dropdown by @NANDINI-star in #12074
• Add Azure OpenAI assistant features cost tracking by @colesmcintosh in #12045
• Remove duplicated entry in logs on key cache update by @Mte90 in #12032
• Update model_prices_and_context_window.json by @codeugar in #11972
• Litellm batch api background cost calc by @krrishdholakia in #12125
• Selecting 'test connect' resets the public model name when selecting an azure model by @NANDINI-star in #11713
• [Bug Fix] Invite links email should contain the correct invite id by @ishaan-jaff in #12130
• fix example config.yaml in claude code tutorial by @glgh in #12133

New Contributors

• @ryan-castner made their first contribution in #12055
• @davis-featherstone made their first contribution in #12075
• @Gum-Joe made their first contribution in #12068
• @jroberts2600 made their first contribution in #12116
• @ohmeow made their first contribution in #12022
• @amarrella made their first contribution in #11942
• @zhangyoufu made their first contribution in #12092
• @bougou made their first contribution in #12088
• @codeugar made their first contribution in #11972
• @glgh made their first contribution in #12133

Full Changelog: v1.73.2-nightly...v1.73.6-nightly

v1.73.6.rc-draft

What's Changed

• Fix SambaNova 'created' field validation error - handle float timestamps by @neubig in #11971
• Docs - Add Recommended Machine Specifications by @ishaan-jaff in #11980
• fix: make response api support Azure Authentication method by @hsuyuming in #11941
• feat: add Last Success column to health check table by @colesmcintosh in #11903
• Add GitHub Actions workflow for LLM translation testing artifacts by @colesmcintosh in #11780
• Fix markdown table not rendering properly by @mukesh-dream11 in #11969
• [Fix] - Check HTTP_PROXY vars in networking requests by @ishaan-jaff in #11947
• Proxy UI MCP Auth passthrough by @wagnerjt in #11968
• fix unrecognised parameter reasoning_effort by @Shankyg in #11838
• Fixing watsonx error: 'model_id' or 'model' cannot be specified in the request body for models in a deployment space by @cbjuan in #11854
• [Bug Fix] Perplexity - LiteLLM doesn't support 'web_search_options' for Perplexity' Sonar Pro model by @ishaan-jaff in #11983
• feat: implement Perplexity citation tokens and search queries cost calculation by @colesmcintosh in #11938
• [Feat] Enterprise - Allow dynamically disabling callbacks in request headers by @ishaan-jaff in #11985
• Add Mistral 3.2 24B to model mapping by @colesmcintosh in #11926
• [Feat] Add List Callbacks API Endpoint by @ishaan-jaff in #11987
• fix: fix test_get_azure_ad_token_with_oidc_token testcase issue by @hsuyuming in #11988
• [Bug Fix] Bedrock Guardrail - Don't raise exception on intervene action by @ishaan-jaff in #11875
• VertexAI Anthropic passthrough cost calc fixes + Filter litellm params from request sent to passthrough endpoint by @krrishdholakia in #11992
• Fix custom pricing logging + Gemini - only use accepted format values + Gemini - cache tools if passing alongside cached content by @krrishdholakia in #11989
• Fix unpack_defs handling of nested $ref inside anyOf items by @colesmcintosh in #11964
• #response_format NVIDIA-NIM add response_format to OpenAI parameters … by @shagunb-acn in #12003
• Add Azure o3-pro Pricing by @marty-sullivan in #11990
• [Bug Fix] SCIM - Ensure new user roles are applied by @ishaan-jaff in #12015
• [Fix] Magistral small system prompt diverges too much from the official recommendation by @ishaan-jaff in #12007
• Refactor unpack_defs to use iterative approach instead of recursion by @colesmcintosh in #12017
• [Feat] Add OpenAI Search Vector Store Operation by @ishaan-jaff in #12018
• [Feat] OpenAI/Azure OpenAI - Add support for creating vector stores on LiteLLM by @ishaan-jaff in #12021
• docs(CLAUDE.md): add development guidance and architecture overview for Claude Code by @colesmcintosh in #12011
• Teams - Support default key expiry + UI - support enforcing access for members of specific SSO Group by @krrishdholakia in #12023
• Anthropic /v1/messages - Custom LLM Server support + Azure Responses api via chat completion support by @krrishdholakia in #12016
• Update mistral 'supports_response_schema' field + Fix ollama embedding by @krrishdholakia in #12024
• [Fix] Router - cooldown time, allow using dynamic cooldown time for a specific deployment by @ishaan-jaff in #12037
• Usage Page: Aggregate the data across all pages by @NANDINI-star in #12033
• [Feat] Add initial endpoints for using Gemini SDK (gemini-cli) with LiteLLM by @ishaan-jaff in #12040
• Add Elasticsearch Logging Tutorial by @colesmcintosh in #11761
• [Feat] Add Support for calling Gemini/Vertex models in their native format by @ishaan-jaff in #12046
• [Feat] Add gemini-cli support - call VertexAI models through LiteLLM Native gemini routes by @ishaan-jaff in #12053
• Managed Files + Batches - filter deployments to only those where file was written + save all model file id mappings in DB (prev just 1st one) by @krrishdholakia in #12048
• Filter team-only models from routing logic for non-team calls + Support List Batches with target model name specified by @krrishdholakia in #12049
• [Feat] gemini-cli integration - Add Logging + Cost tracking for stream + non-stream Vertex / Google AI Studio routes by @ishaan-jaff in #12058
• Fix Elasticsearch tutorial image rendering by @colesmcintosh in #12050
• [Fix] Allow using HTTP_ Proxy settings with trust_env by @ishaan-jaff in #12066
• fix(proxy): Fix test_mock_create_audio_file by adding managed_files hook by @colesmcintosh in #12072
• Enhance CircleCI integration in LLM translation testing workflow by @colesmcintosh in #12041
• Inkeep searchbar and chat added to the Docs by @NANDINI-star in #12030
• [Fix] Redis - Add better debugging to see what variables are set by @ishaan-jaff in #12073
• Fix today selector date mutation bug in dashboard components by @colesmcintosh in #12042
• Responses API - Add reasoning content support for non-OpenAI providers by @ryan-castner in #12055
• Litellm dev 06 26 2025 p1 by @krrishdholakia in #12087
• Refactor: bedrock passthrough fixes - migrate to Passthrough SDK by @krrishdholakia in #12089
• Fix Azure-OpenAI Vision API Compliance by @davis-featherstone in #12075
• [Bug Fix] Bedrock Guardrails - Ensure PII Masking is applied on response streaming or non streaming content when using post call by @ishaan-jaff in #12086
• fix(docs): Remove unused dotenv dependency from docusaurus config by @colesmcintosh in #12102
• [Fix] MCP - Ensure internal users can access /mcp and /mcp/ routes by @ishaan-jaff in #12106
• fix: handle provider_config type error in passthrough error handler by @colesmcintosh in #12101
• Add o3 and o4-mini deep research models by @krrishdholakia in #12109
• [Bug Fix] Anthropic - Token Usage Null Handling in calculate_usage by @Gum-Joe in #12068
• fix: change cost calculation logs from INFO to DEBUG level by @colesmcintosh in #12112
• fix: set logger levels based on LITELLM_LOG environment variable by @colesmcintosh in #12111
• [Feat] Add Bridge from generateContent <> /chat/completions by @ishaan-jaff in #12081
• [Docs] - Show how to use fallbacks with audio transcriptions endpoints by @ishaan-jaff in #12115
• [Bug Fix] Fix handling str, bool types formock_testing_fallbacks on router using /audio endpoints by @ishaan-jaff in #12117
• Adding Feature: Palo Alto Networks Prisma AIRS Guardrail by @jroberts2600 in #12116
• [Bug Fix] Exception mapping for context window exceeded - should catch anthropic exceptions by @ishaan-jaff in #12113
• docs(GEMINI.md): add development guidelines and architecture overview by @colesmcintosh in #12035
• [Bug fix] Router - handle cooldown_time = 0 for deployments by @ishaan-jaff in #12108
• [Feat] Add Eleven Labs - Speech To Text Support on LiteLLM by @ishaan-jaff in #12119
• Revert "fix: set logger levels based on LITELLM_LOG environment variable" by @ishaan-jaff in #12122
• Fix Braintrust integration: Adds model to metadata to calculate cost and corrects docs by @ohmeow in #12022
• [Fix] Change Message init type annotation to support other roles by @amarrella in #11942
• Add "Get Code" Feature by @NANDINI-star in #11629
• Bedrock Passthrough cost tracking (/invoke + /converse routes - streaming + non-streaming) by @krrishdholakia in #12123
• feat: ad...

v1.73.2-nightly

What's Changed

• VertexAI Anthropic passthrough cost calc fixes + Filter litellm params from request sent to passthrough endpoint by @krrishdholakia in #11992
• Fix custom pricing logging + Gemini - only use accepted format values + Gemini - cache tools if passing alongside cached content by @krrishdholakia in #11989
• Fix unpack_defs handling of nested $ref inside anyOf items by @colesmcintosh in #11964
• #response_format NVIDIA-NIM add response_format to OpenAI parameters … by @shagunb-acn in #12003
• Add Azure o3-pro Pricing by @marty-sullivan in #11990
• [Bug Fix] SCIM - Ensure new user roles are applied by @ishaan-jaff in #12015
• [Fix] Magistral small system prompt diverges too much from the official recommendation by @ishaan-jaff in #12007
• Refactor unpack_defs to use iterative approach instead of recursion by @colesmcintosh in #12017
• [Feat] Add OpenAI Search Vector Store Operation by @ishaan-jaff in #12018
• [Feat] OpenAI/Azure OpenAI - Add support for creating vector stores on LiteLLM by @ishaan-jaff in #12021
• docs(CLAUDE.md): add development guidance and architecture overview for Claude Code by @colesmcintosh in #12011
• Teams - Support default key expiry + UI - support enforcing access for members of specific SSO Group by @krrishdholakia in #12023
• Anthropic /v1/messages - Custom LLM Server support + Azure Responses api via chat completion support by @krrishdholakia in #12016
• Update mistral 'supports_response_schema' field + Fix ollama embedding by @krrishdholakia in #12024
• [Fix] Router - cooldown time, allow using dynamic cooldown time for a specific deployment by @ishaan-jaff in #12037
• Usage Page: Aggregate the data across all pages by @NANDINI-star in #12033
• [Feat] Add initial endpoints for using Gemini SDK (gemini-cli) with LiteLLM by @ishaan-jaff in #12040
• Add Elasticsearch Logging Tutorial by @colesmcintosh in #11761
• [Feat] Add Support for calling Gemini/Vertex models in their native format by @ishaan-jaff in #12046
• [Feat] Add gemini-cli support - call VertexAI models through LiteLLM Native gemini routes by @ishaan-jaff in #12053
• Managed Files + Batches - filter deployments to only those where file was written + save all model file id mappings in DB (prev just 1st one) by @krrishdholakia in #12048
• Filter team-only models from routing logic for non-team calls + Support List Batches with target model name specified by @krrishdholakia in #12049
• [Feat] gemini-cli integration - Add Logging + Cost tracking for stream + non-stream Vertex / Google AI Studio routes by @ishaan-jaff in #12058
• Fix Elasticsearch tutorial image rendering by @colesmcintosh in #12050
• [Fix] Allow using HTTP_ Proxy settings with trust_env by @ishaan-jaff in #12066

Full Changelog: v1.73.1-nightly...v1.73.2-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.2-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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	210.0	227.43264090448105	6.195122505342355	0.0	1853	0	184.98149800007013	1558.9818880000053
Aggregated	Passed ✅	210.0	227.43264090448105	6.195122505342355	0.0	1853	0	184.98149800007013	1558.9818880000053

v1.73.0-stable

Full Changelog: v1.73.0.rc.1...v1.73.0-stable

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:litellm_stable_release_branch-v1.73.0-stable

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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	230.0	247.8644959408947	6.280622823577799	0.003344314602544089	1878	1	201.14030100000946	1471.8547979999812
Aggregated	Passed ✅	230.0	247.8644959408947	6.280622823577799	0.003344314602544089	1878	1	201.14030100000946	1471.8547979999812

v1.73.2.dev1

What's Changed

• VertexAI Anthropic passthrough cost calc fixes + Filter litellm params from request sent to passthrough endpoint by @krrishdholakia in #11992
• Fix custom pricing logging + Gemini - only use accepted format values + Gemini - cache tools if passing alongside cached content by @krrishdholakia in #11989
• Fix unpack_defs handling of nested $ref inside anyOf items by @colesmcintosh in #11964
• #response_format NVIDIA-NIM add response_format to OpenAI parameters … by @shagunb-acn in #12003
• Add Azure o3-pro Pricing by @marty-sullivan in #11990
• [Bug Fix] SCIM - Ensure new user roles are applied by @ishaan-jaff in #12015

Full Changelog: v1.73.1-nightly...v1.73.2.dev1

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.2.dev1

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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	250.0	267.8382003869747	6.2096771619800935	0.0	1858	0	214.47131599995828	1466.6541370000346
Aggregated	Passed ✅	250.0	267.8382003869747	6.2096771619800935	0.0	1858	0	214.47131599995828	1466.6541370000346

v1.73.1-nightly

What's Changed

• Fix SambaNova 'created' field validation error - handle float timestamps by @neubig in #11971
• Docs - Add Recommended Machine Specifications by @ishaan-jaff in #11980
• fix: make response api support Azure Authentication method by @hsuyuming in #11941
• feat: add Last Success column to health check table by @colesmcintosh in #11903
• Add GitHub Actions workflow for LLM translation testing artifacts by @colesmcintosh in #11780
• Fix markdown table not rendering properly by @mukesh-dream11 in #11969
• [Fix] - Check HTTP_PROXY vars in networking requests by @ishaan-jaff in #11947
• Proxy UI MCP Auth passthrough by @wagnerjt in #11968
• fix unrecognised parameter reasoning_effort by @Shankyg in #11838
• Fixing watsonx error: 'model_id' or 'model' cannot be specified in the request body for models in a deployment space by @cbjuan in #11854
• [Bug Fix] Perplexity - LiteLLM doesn't support 'web_search_options' for Perplexity' Sonar Pro model by @ishaan-jaff in #11983
• feat: implement Perplexity citation tokens and search queries cost calculation by @colesmcintosh in #11938
• [Feat] Enterprise - Allow dynamically disabling callbacks in request headers by @ishaan-jaff in #11985
• Add Mistral 3.2 24B to model mapping by @colesmcintosh in #11926
• [Feat] Add List Callbacks API Endpoint by @ishaan-jaff in #11987
• fix: fix test_get_azure_ad_token_with_oidc_token testcase issue by @hsuyuming in #11988
• [Bug Fix] Bedrock Guardrail - Don't raise exception on intervene action by @ishaan-jaff in #11875

New Contributors

• @mukesh-dream11 made their first contribution in #11969
• @cbjuan made their first contribution in #11854

Full Changelog: v1.73.0.rc.1...v1.73.1-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.1-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

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.1-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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	250.0	269.80153099125215	6.123419901585826	0.0	1829	0	217.6905329999954	1336.1768169999948
Aggregated	Passed ✅	250.0	269.80153099125215	6.123419901585826	0.0	1829	0	217.6905329999954	1336.1768169999948

v1.73.0.rc.1

What's Changed

• (Tutorial) Onboard Users for AI Exploration by @krrishdholakia in #11955
• Management Fixes - don't apply default internal user settings to admins + preserve all model access for teams with empty model list, when team model added + /v2/model/info fixes by @krrishdholakia in #11957

Full Changelog: v1.73.0-nightly...v1.73.0.rc.1

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.0.rc.1

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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	220.0	240.58426726332922	6.145106667198675	0.0	1838	0	196.0181700000021	1838.010895000025
Aggregated	Passed ✅	220.0	240.58426726332922	6.145106667198675	0.0	1838	0	196.0181700000021	1838.010895000025

v1.73.0-nightly

What's Changed

• Update Azure o3 pricing to match OpenAI pricing ($2/$8 per 1M tokens) by @ervwalter in #11937
• [BugFix] Ollama response_format not working by @ThakeeNathees in #11880
• fix aws bedrock claude tool call index by @jnhyperion in #11842
• fix(acompletion): allow dict for tool_choice argument by @Jannchie in #11860
• [Chore] Check team counts on license when creating new team by @ishaan-jaff in #11943
• [Docs] [Pre-Release] v1.73.0-stable by @ishaan-jaff in #11950
• Show user all models they can call (Across teams) on UI by @krrishdholakia in #11948

New Contributors

• @ervwalter made their first contribution in #11937
• @ThakeeNathees made their first contribution in #11880
• @jnhyperion made their first contribution in #11842
• @Jannchie made their first contribution in #11860

Full Changelog: v1.72.9-nightly...v1.73.0-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.0-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

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.73.0-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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	230.0	252.55233764162196	6.182759830384375	0.0	1850	0	208.6453730000244	1743.1928639999796
Aggregated	Passed ✅	230.0	252.55233764162196	6.182759830384375	0.0	1850	0	208.6453730000244	1743.1928639999796

v1.72.9-nightly

What's Changed

• [Feat] MCP - Allow connecting to MCP with authentication headers + Allow clients to specify MCP headers (#11890) by @ishaan-jaff in #11891
• [Fix] Networking - allow using CA Bundles by @ishaan-jaff in #11906
• [Feat] Add AWS Bedrock profiles for the APAC region by @lgruen-vcgs in #11883
• bumps the anthropic package by @rinormaloku in #11851
• Add deployment annotations by @InvisibleMan1306 in #11849
• Enhance Mistral API: Add support for parallel tool calls by @njbrake in #11770
• [UI] QA Items for adding pass through endpoints by @ishaan-jaff in #11909
• build(model_prices_and_context_window.json): mark all gemini-2.5 models support pdf input + Set anthropic custom llm provider property by @krrishdholakia in #11907
• fix(proxy_server.py): fix loading ui on custom root path by @krrishdholakia in #11912
• LiteLLM SDK <-> Proxy improvement (don't transform message client-side) + Bedrock - handle qs:.. in base64 file data + Tag Management - support adding public model names by @krrishdholakia in #11908
• Add success modal for health check responses by @colesmcintosh in #11899
• Volcengine - thinking param support + Azure - handle more gpt custom naming patterns by @krrishdholakia in #11914
• [Feat] Model Cost Map - Add gemini-2.5-pro and set gemini-2.5-pro supports_reasoning=True by @ishaan-jaff in #11927
• [Feat] UI Allow testing /v1/messages on the Test Key Page by @ishaan-jaff in #11930
• Feat/add delete callback by @jtong99 in #11654
• add ciphers in command and pass to hypercorn for proxy by @frankzye in #11916
• [Bug Fix] Fix model_group tracked for /v1/messages and /moderations by @ishaan-jaff in #11933
• [Bug Fix] Cost tracking and logging via the /v1/messages API are not working when using Claude Code by @ishaan-jaff in #11928
• [Feat] Add Azure Codex Models on LiteLLM + new /v1 preview Azure OpenAI API by @ishaan-jaff in #11934
• [Feat] UI QA: Pass through endpoints by @ishaan-jaff in #11939

New Contributors

• @lgruen-vcgs made their first contribution in #11883
• @rinormaloku made their first contribution in #11851
• @InvisibleMan1306 made their first contribution in #11849

Full Changelog: v1.72.7-nightly...v1.72.9-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.72.9-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

Name	Status	Median Response Time (ms)	Average Response Time (ms)	Requests/s	Failures/s	Request Count	Failure Count	Min Response Time (ms)	Max Response Time (ms)
/chat/completions	Passed ✅	240.0	259.91299670392254	6.2187422072270495	0.0	1861	0	210.9276310000041	1676.9406920000165
Aggregated	Passed ✅	240.0	259.91299670392254	6.2187422072270495	0.0	1861	0	210.9276310000041	1676.9406920000165