scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-09 03:18:44 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 109
24472 commits 3347 branches 1448 tags 5 GiB
Commit graph

24472 commits

This branch
This branch
All branches
Author SHA1 Message Date
Ishaan Jaff
3a76b2762f ci/cd new release 2025-07-22 20:41:49 -07:00
Ishaan Jaff
5741cdae0f fix openrouter/qwen/qwen-vl-plus 2025-07-22 20:37:57 -07:00
xywei
ed0ad6fd59 Help mypy with typing asearch 2025-07-22 22:18:02 -05:00
Krish Dholakia
8cd6c25e1e
Fix async retryer on .acompletion() + forward clientside headers - filter out content-type from clientside request (causes llm api call to hang) (#12886) 
* fix(main.py): fix async retryer

Fixes https://github.com/BerriAI/litellm/issues/12830

* fix(forward_clientside_headers_by_model_group.py): filter out 'content-type' from forwardable headers

clientside content-type != proxy content type, can cause requests to hang
 2025-07-22 19:50:05 -07:00
xywei
f81161a05c Remove vector store methods from global scope 2025-07-22 21:16:39 -05:00
Ishaan Jaff
b41ce5c92f
[Feat] - Track cost + add tags for health checks done by LiteLLM Proxy (#12880) 
* refactor to use add_user_api_key_auth_to_request_metadata

* add get_litellm_internal_health_check_user_api_key_auth

* add get_litellm_internal_health_check_user_api_key_auth

* add _update_model_params_with_health_check_tracking_information

* add HealthCheckHelpers

* refactor to use clean helpers

* test_update_model_params_with_health_check_tracking_information

* test_get_litellm_internal_health_check_user_api_key_auth

* test_add_user_api_key_auth_to_request_metadata

* fix _update_model_params_with_health_check_tracking_information
 2025-07-22 18:45:57 -07:00
Ishaan Jaff
c21dc46a33 fix morph api tests 2025-07-22 18:44:44 -07:00
Ishaan Jaff
bd8b21e700 fix mypy linting 2025-07-22 18:38:54 -07:00
Ishaan Jaff
d93e85282e test_bad_database_url 2025-07-22 18:37:51 -07:00
Ishaan Jaff
bf300f8ca7 Revert "Litellm dev 07 21 2025 p1 (#12848)" 
This reverts commit e4e10aa4ed.
 2025-07-22 18:28:36 -07:00
Ishaan Jaff
f0a8abb911 fix cost_calculator 2025-07-22 18:16:53 -07:00
Ishaan Jaff
4b741adadb fix recraft cost calc 2025-07-22 18:12:12 -07:00
Ishaan Jaff
e5debfc84c fix model cost map for recraft 2025-07-22 18:11:10 -07:00
Ishaan Jaff
1910cf8496 test fix vertex ai 2025-07-22 18:06:38 -07:00
Ishaan Jaff
b026f6b280
[Feat] Add cost tracking for new vertex_ai/llama-3 API models (#12878) 
* add vertex_ai/meta/llama-3.1-405b-instruct-maas

* notes - vertex_ai/meta/llama-3.2-90b-vision-instruct-maas
 2025-07-22 16:22:09 -07:00
Jugal D. Bhatt
c9899b5c06
[LLM Translation] Litellm gemini 2.0 live support (#12839) 
* add gemini 2.0 live to model context and priceS

* added files to dump

* Update litellm/model_prices_and_context_window_backup.json

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update model_prices_and_context_window.json

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* added vertex ai live preview

* added vertex ai change

* add input video and image cosT

* add input video and image cosT

---------

Co-authored-by: Ishaan Jaff <ishaanjaffer0324@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
 2025-07-22 15:14:52 -07:00
Ishaan Jaff
d5ee93aa0c
[Feat] Add Recraft API - Image Edits Support (#12874) 
* test_recraft_image_edit_api

* add RecraftImageEditConfig

* complete RecraftImageEditConfig

* add RecraftImageEditRequestParams in types

* update RecraftImageEditRequestParams

* working

* transform_image_edit_request

* Image Edit docs recraft

* working transform_image_edit_request

* TestRecraftImageEditTransformation
 2025-07-22 15:03:08 -07:00
Ishaan Jaff
31e9303232 remove old test 2025-07-22 14:11:43 -07:00
Jugal D. Bhatt
aa1ebf37b2
pass in cmd args (#12871) 2025-07-22 14:10:23 -07:00
Ishaan Jaff
934af0e9a0
add azure-keyvault==4.2.0 (#12873) 2025-07-22 14:08:33 -07:00
dependabot[bot]
d4900d7dc0
build(deps): bump form-data from 4.0.3 to 4.0.4 in /docs/my-website (#12867) 
---
updated-dependencies:
- dependency-name: form-data
  dependency-version: 4.0.4
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2025-07-22 11:23:41 -07:00
Ishaan Jaff
6f0c211788 [Feat] MCP Gateway - allow using MCPs with all LLM APIs when using /responses with LiteLLM (#12546) 
* add MCPResponsesAPIHelper

* rename LiteLLM_Proxy_MCP_Handler

* aresponses_api_with_mcp

* mock_responses_api_response

* test response with litellm proxy MCP

* add _should_use_litellm_mcp_gateway

* fix transform_mcp_tool_to_openai_responses_api_tool

* use correct _transform_mcp_tools_to_openai

* fix config.yaml

* fixes for native MCP handling

* docs MCP with litellm proxy

* aresponses_api_with_mcp

* fix linting

* fix mypy

* fix linting

* test_aresponses_api_with_mcp_mock_integration

* docs How it works when server_url="litellm_proxy"
 2025-07-22 10:22:45 -07:00
Krrish Dholakia
e683a41993 bump: version 1.74.7 → 1.74.8 2025-07-22 08:54:03 -07:00
Mateo Di Loreto
c65392cf81
Replace non-root Dockerfile base with Alpine multi-stage build; (#12707) 
* Change Dockerfile.noon_root with alpine base image

* Improve non_root docker image

* Re add the build_admin_ui.sh script step

* Re add the build_admin_ui.sh script step

* Remove unnecessary workdir set

* Remove unnecessary workdir set

* Configure chainguard image

* A bit of optimization and improve comments

* delete extra build_ui script run

* Optimizie Dockerfile copy statements
 2025-07-22 08:53:10 -07:00
Krrish Dholakia
964c3b7a3d docs(index.md): cleanup 2025-07-22 08:47:52 -07:00
Krrish Dholakia
b4469dafd4 docs(index.md): cleanup 2025-07-22 08:47:26 -07:00
Krrish Dholakia
581e96a768 docs(index.md): cleanup 2025-07-22 08:43:33 -07:00
Adam Holmberg
2ce03d9735 fix: make gemini and openai responses return reasoning by default 
This aligns the proxy experience with other models that think
automatically (e.g. Deepseek R1 and grok3). It does so by setting
the necessary request input to return thinking, but not specifying
a budget or effort (thus defaulting to the internal automatic level).
 2025-07-22 10:41:29 -05:00
tanjiro
03e420f91a
Improvements on the Regenerate Key Flow (#12788) 
* improve regeneration ux

* toLocaleString

* regenerate key twice in a row

* fix delete key
 2025-07-22 07:47:26 -07:00
Ishaan Jaff
03baf23ad1 [Feat] Add Recraft Image Generation API Support - New LLM Provider (#12832) 
* add recraft

* init RecraftImageGenerationConfig

* add get_complete_url + validate_environment

* add image_generation_handler in llm http clas

* fixes for transform

* working recraft request

* fixed img gen transform

* fixes for llm http handler

* test: TestRecraftImageGeneration

* fixes for llm_http_handler

* fix RecraftImageGenerationConfig

* TestRecraftImageGenerationTransformation

* add recraft API

* docs recraft API

* fix code QA

* map_openai_params

* fix recraft

* cost tracking for recraft/recraftv3

* fix code qa check
 2025-07-21 22:19:58 -07:00
Krish Dholakia
e5251e7188
Openrouter - filter out cache_control flag for non-anthropic models (allows usage with claude code) (#12850) 
* fix(gpt_transformation.py): remove 'cache_control' flag for openai/openai-compatible calls

Fixes https://github.com/BerriAI/litellm/issues/12787

* fix(openrouter/chat/transformation.py): allow passing openrouter cache control flag for claude models

* fix(gpt_transformation.py): fix import

* fix: fix adding tools
 2025-07-21 22:15:48 -07:00
Krish Dholakia
e4e10aa4ed
Litellm dev 07 21 2025 p1 (#12848) 
* fix(main.py): fix async retryer

Fixes https://github.com/BerriAI/litellm/issues/12830

* fix(forward_clientside_headers_by_model_group.py): filter out 'content-type' from forwardable headers

clientside content-type != proxy content type, can cause requests to hang

* test(tests/): update tests
 2025-07-21 22:09:39 -07:00
Krish Dholakia
db498c1805
Fix team_member_budget update logic (#12843) 
* fix(team_endpoints.py): always remove team member budget from updated_kv

this is not a field for the litellm team table

Prevents startup issue

* test(test_team_endpoints.py): add unit test to ensure 'team_member_budget' is never in update to table - separate logic

* refactor: cleanup
 2025-07-21 22:06:29 -07:00
dependabot[bot]
2be16eb493
build(deps): bump form-data from 4.0.0 to 4.0.4 in /ui/litellm-dashboard (#12851) 
---
updated-dependencies:
- dependency-name: form-data
  dependency-version: 4.0.4
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2025-07-21 22:05:38 -07:00
Krish Dholakia
70ffdcd210
Passthrough Auth - make Auth checks OSS + Anthropic - only show 'reasoning_effort' for supported models (#12847) 
* fix(anthropic/chat/transformation.py): move 'reasoning_effort' to inside 'supports_reasoning' check

Fixes https://github.com/BerriAI/litellm/issues/12833

* fix(proxy_settings_endpoints.py): change order of checks - don't add allowed ip unless db is connected

Fixes https://github.com/BerriAI/litellm/issues/12815

* feat(user_api_key_auth.py): make passthrough auth checks OSS

Fixes https://github.com/BerriAI/litellm/issues/12789

* docs(enterprise.md): clarify on docs
 2025-07-21 21:55:41 -07:00
Ishaan Jaff
49d40a1c3d test_router_provider_wildcard_routing 2025-07-21 21:33:40 -07:00
Krrish Dholakia
68a5415a51 docs(config_settings.md): document new flags 2025-07-21 20:19:26 -07:00
Krish Dholakia
00677afefe
Litellm batch cost tracking debug (#12782) 
* feat(proxy_server.py): support batch polling interval

allows admin to control batch polling interval (default is 3600s)

easier debugging

* fix(proxy_settings_endpoint.py): ensure value is actually set before updating env var
 2025-07-21 20:17:56 -07:00
Cole McIntosh
ff22aed1ea
Merge pull request #12826 from colesmcintosh/feature/add-hyperbolic-provider 2025-07-21 20:10:45 -06:00
Tomáš Dvořák
270e3d75db
fix(watsonx): use correct parameter name for tool choice (#9980) 
Closes BerriAI/litellm#9979
 2025-07-21 19:01:10 -07:00
Ishaan Jaff
c91712e579 bump: version 1.74.7 → 1.74.8 2025-07-21 18:29:34 -07:00
Ishaan Jaff
83463e3ea5 fix docusaurus package lock 2025-07-21 18:29:17 -07:00
Ishaan Jaff
443e26f46d fix openrouter/qwen/qwen-vl-plus 2025-07-21 18:25:25 -07:00
Ishaan Jaff
4a7b9dee5f test fix - anthropic deprecated claude 2 2025-07-21 18:22:39 -07:00
Ishaan Jaff
4ca3e8e617
Docs - litellm benchmarks (#12842) 
* docs litellm overhead

* In these tests the baseline latency characteristics
 2025-07-21 18:10:30 -07:00
Ishaan Jaff
133c26c015
[Azure OpenAI Feature] - Support DefaultAzureCredential without hard-coded environment variables (#12841) 
* DefaultAzureCredential

* update get_azure_ad_token_provider

* fixes for get_azure_ad_token_provider

* test_get_azure_ad_token_provider_with_default_azure_credential

* test_get_azure_ad_token_fallback_to_default_azure_credential

* docs DefaultAzureCredential

* fix linting
 2025-07-21 18:04:16 -07:00
Jugal D. Bhatt
7bdb5593bf
[LLM Translation] add qwen-vl-plus (#12829) 
* add qwen-vl-plus

* add qwen-vl-plus
 2025-07-21 16:58:59 -07:00
Cole McIntosh
34ccda10ed Merge upstream/main - resolve conflicts to include both hyperbolic and recraft providers 2025-07-21 17:38:17 -06:00
Ishaan Jaff
27c9be67ba
[Feat] Add fireworks - fireworks/models/kimi-k2-instruct (#12837) 
* add fireworks - fireworks/models/kimi-k2-instruct

* update source
 2025-07-21 16:28:02 -07:00
Ishaan Jaff
1b05ea79ce update docs 2025-07-21 15:52:54 -07:00