scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-08 03:08:45 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 115
26247 commits 3285 branches 1443 tags 4.8 GiB
Commit graph

16050 commits

Author SHA1 Message Date
Tim Elfrink
56b2c914ed Apply isort formatting to router.py imports 2025-10-09 08:13:16 +02:00
Tim Elfrink
02f0b1e83a Merge latest changes from origin/main into fix/redact-aws-credentials 
Resolved import conflict for CustomPricingLiteLLMParams which was moved
from litellm.types.router to litellm.types.utils in main branch.
 2025-10-09 08:12:22 +02:00
Krish Dholakia
e2e0cdde31
Merge pull request #15315 from lcfyi/lcfyi/add-support-for-parallel-tool-calls 
Fix parallel tool calls in the Anthropic passthrough adapter
 2025-10-08 23:01:55 -07:00
Krish Dholakia
4615298ba1
Merge pull request #15336 from BerriAI/litellm_fix_litellm_Param_costing 
Fix litellm_param based costing
 2025-10-08 22:55:41 -07:00
Sameer Kankute
e4b89813a5 Add remaining cost fields 2025-10-09 08:40:06 +05:30
Ishaan Jaffer
763d2f8ccd UI new build 2025-10-08 19:57:33 -07:00
Krish Dholakia
a02b3caef3
Merge branch 'main' into litellm_dev_10_08_2025_p2 2025-10-08 19:27:24 -07:00
Krish Dholakia
d921df1baf
Merge pull request #15343 from BerriAI/litellm_dev_10_08_2025_p1 
MCP - support converting OpenAPI specs to MCP servers
 2025-10-08 19:10:04 -07:00
Krrish Dholakia
5be198248b fix: fix linting error 2025-10-08 19:07:11 -07:00
Krrish Dholakia
37fdb486d4 refactor: comment out oauth error raising logic for now 2025-10-08 19:06:19 -07:00
Krish Dholakia
8cbc85578e
Merge pull request #15351 from BerriAI/litellm_dev_10_08_2025_p3 
SSO - support EntraID app roles
 2025-10-08 19:04:00 -07:00
Krish Dholakia
12a1d081ee
Merge branch 'main' into litellm_dev_09_11_2025_p1 2025-10-08 19:02:58 -07:00
Krrish Dholakia
8eafb7a8fa docs: doc improvements 2025-10-08 18:50:42 -07:00
Krrish Dholakia
d28ecbc900 feat(ui_sso.py): support mapping app roles from azure entra id to litellm user roles 
Closes LIT-1228
 2025-10-08 18:45:13 -07:00
Ishaan Jaffer
c4022ade49 test mapped tests MCP 2025-10-08 18:34:30 -07:00
Copilot
4226314096
Add native Responses API support for litellm_proxy provider (#15347) 
* Initial plan

* Add native Responses API support for litellm_proxy provider

Co-authored-by: ishaan-jaff <29436595+ishaan-jaff@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: ishaan-jaff <29436595+ishaan-jaff@users.noreply.github.com>
 2025-10-08 18:31:26 -07:00
Ishaan Jaff
2f42c806cb
[Fix] x-litellm-cache-key header not being returned on cache hit (#15348) 
* fix: x-cache-key

* test_cache_key_in_hidden_params_acompletion

* fix: remove_cache_control_flag_from_messages_and_tools
 2025-10-08 18:10:43 -07:00
Ishaan Jaff
97031dc8ee
Fix - (openrouter): move cache_control to content blocks for claude/gemini (#15345) 
* test_openrouter_transform_request_with_cache_control

* fix CacheControlSupportedModels

* test_openrouter_transform_request_with_cache_control_gemini
 2025-10-08 17:41:04 -07:00
Krrish Dholakia
c97495c5d9 docs(mcp.md): document new allow/disallow tool parameters functionality 2025-10-08 17:39:42 -07:00
Krrish Dholakia
0b27c361fe feat(mcp/): initial commit raising correct oauth error 2025-10-08 17:33:11 -07:00
Pierre-Emmanuel MERCIER
861750790b feat: add redis ssl and username support (#11319) 2025-10-08 16:48:07 -07:00
Krrish Dholakia
ecb6c0e814 fix(mcp_server_manager.py): ensure only allowed params sent to MCP server 2025-10-08 16:37:50 -07:00
Krrish Dholakia
dd4249db78 feat(mcp_server_manager.py): initial commit adding allowed params support 
allow admin to specify which parameters to allow/disallow by MCP tool
 2025-10-08 16:27:01 -07:00
Krrish Dholakia
cd13e533a2 feat(mcp_server_manager.py): expand allowed/disallowed list 2025-10-08 16:06:17 -07:00
Krrish Dholakia
28698933fc feat(mcp_server_manager.py): ensure list tools for openapi servers works as expected 2025-10-08 15:59:42 -07:00
Ishaan Jaff
1c56a0d856
[Fix] Watsonx - Apply correct prompt templates for openai/gpt-oss model family (#15341) 
* fix: apply_prompt_template

* Revert "fix: apply_prompt_template"

This reverts commit 3e0e40b497.

* add apply_prompt_template for WatsonX

* feat: add apply_prompt_template

* test_watsonx_gpt_oss_prompt_transformation

* Revert "add apply_prompt_template for WatsonX"

This reverts commit 3e80903796.

* add apply_prompt_template for WatsonX

* fix apply_prompt_template

* fix: add hf template handler

* fix hf_chat_template

* fix _get_tokenizer_config

* fix hf_chat_template

* add WatsonXModelPattern

* fix aapply_prompt_template
 2025-10-08 15:39:36 -07:00
Krrish Dholakia
1c317a35af feat(mcp/): initial working list tools for openapi spec mcp tools 2025-10-08 15:32:31 -07:00
Krrish Dholakia
ff893cce42 feat(mcp_server_manager.py): working call tools for openapi to mcp servers 2025-10-08 15:31:52 -07:00
Krrish Dholakia
386468ad77 feat: initial commit for openapi to mcp conversion 
allows users to give openapi spec for a server, and litellm to serve it as an mcp
 2025-10-08 15:27:48 -07:00
Ishaan Jaffer
9d84a7cc8b Revert "fix: apply_prompt_template" 
This reverts commit 3e0e40b497.
 2025-10-08 13:56:47 -07:00
Ishaan Jaffer
3e0e40b497 fix: apply_prompt_template 2025-10-08 13:56:25 -07:00
Felipe Garé
f12b8bf2c1
feat(files): add @client decorator to file operations (#15339) 
- Decorate sync/async file APIs (retrieve, delete, list, content) with @client
- Ensures calls route through the configured client, honoring provider/client config
- Standardizes behavior across file endpoints and aligns with existing patterns
- No API signature changes; improves consistency and client-bound usage
 2025-10-08 13:42:34 -07:00
Krrish Dholakia
45ec1ffdaa fix(ui/): set is sso modal visible to true for all users 
allow backend to handle more complex state management
 2025-10-08 13:13:21 -07:00
Sameer Kankute
0b6b69cd5b
fix issue with parsing assistant messages (#15320) 
* Update factory.py

* reinclude the test cases

* fixed test

---------

Co-authored-by: Weijie <weijie-tan3@github.com>
 2025-10-08 09:49:02 -07:00
Sameer Kankute
85d4142845 Fix litellm_param based costing 2025-10-08 21:14:23 +05:30
Tim Elfrink
f8af72bbf1 fix: redact AWS credentials when redact_user_api_key_info enabled 
Use SensitiveDataMasker in print_deployment() to mask all sensitive
credentials including AWS keys when redact_user_api_key_info is True.

Fixes #14839
 2025-10-08 10:31:35 +02:00
Leslie Cheng
86affa405c Unused type 2025-10-07 21:11:56 -07:00
Leslie Cheng
1579752e06 Implement fix 2025-10-07 20:53:38 -07:00
Krish Dholakia
13703f289b
Merge pull request #15292 from timelfrink/fix/bedrock-prompt-caching-cost-calculation 
fix(bedrock): include cacheWriteInputTokens in prompt_tokens calculation
 2025-10-07 19:11:45 -07:00
Krish Dholakia
3f4d2c6ada
Add Cohere Embed v4 support for AWS Bedrock 
Add Cohere Embed v4 support for AWS Bedrock
 2025-10-07 19:10:22 -07:00
Ishaan Jaffer
5d6cd3ffed UI new build 2025-10-07 18:59:11 -07:00
Ishaan Jaffer
a8243079d3 doc fix 2025-10-07 18:51:27 -07:00
Ishaan Jaff
8a8cc5a1d3
[QA/Fixes] - Dynamic Rate Limiter v3 - final QA (#15311) 
* fix: PriorityReservationSettings

* fix: use correct casting

* fix type casting
 2025-10-07 18:42:17 -07:00
Ishaan Jaff
36c971a6fd
[MCP Gateway] QA/Fixes - Ensure Team/Key level enforcement works for MCPs (#15305) 
* fix: _set_object_permission

* fix: _set_object_permission on teams

* fix: _set_object_permission

* fixes for team/key permissions

* statsh: object permission view

* fix: MCPServerPermissions

* fix: _get_team_object_permission

* test mcp checks for permissions

* fix server checks with prefix names

* test_list_tools_strips_prefix_when_matching_permissions

* ruff fix

* docs - refactor MCP

* docs update MCP docs

* docs allowed tools
 2025-10-07 17:34:48 -07:00
Ishaan Jaff
7b56ba240e
[MCP Gateway] Litellm mcp fixes team control (#15304) 
* fix: _set_object_permission

* fix: _set_object_permission on teams

* fix: _set_object_permission

* fixes for team/key permissions

* statsh: object permission view

* fix: MCPServerPermissions
 2025-10-07 16:48:00 -07:00
Alexsander Hamir
49e04e0217
[Fix] Networking: remove limitations (#15302) 
* fix: remove limitations

* fix: linter issues
 2025-10-07 16:45:42 -07:00
Ishaan Jaff
bc26eff98f
Fix: Make PATCH /model/{model_id}/update handle team_id consistently with POST /model/new (#15297) 
* fix: _update_team_model_in_db

* test_patch_model_with_team_id_creates_proper_setup
 2025-10-07 14:04:08 -07:00
Tim Elfrink
d71d801e4d Add Cohere Embed v4 support for AWS Bedrock 
- Add cohere.embed-v4:0 to model pricing configs
- Update bedrock_embedding_models constant
- Update documentation with v4 model support

Fixes #15272
 2025-10-07 22:08:12 +02:00
Ishaan Jaff
07a17d6d6b
[Feat] Proxy CLI - dont store existing key in the URL, store it in the state param (#15290) 
* Feat: CLI Auth fixes for UI SSO

* fix auth.py

* fix test ui sso.py
 2025-10-07 12:36:17 -07:00
Tim Elfrink
c5eb22381d fix(bedrock): include cacheWriteInputTokens in prompt_tokens calculation 
Fixes #15263

This PR fixes the cost calculation for Bedrock Anthropic models with prompt caching.

**Root Cause:**
PR #9838 incorrectly removed adding `cacheWriteInputTokens` to `prompt_tokens`
for Bedrock, based on the assumption that it would cause double counting (similar
to an Anthropic API issue). However, Bedrock's token structure is different:

- **Bedrock API**: `inputTokens`, `cacheReadInputTokens`, and `cacheWriteInputTokens`
  are ALL separate values that should be summed for total input tokens
- **Anthropic API**: Same structure - all three token types are separate

The fix in #9838 was later reverted for Anthropic (correctly re-adding
`cache_creation_input_tokens` to `prompt_tokens`), but Bedrock was never fixed.

**Changes:**
1. Re-add `cacheWriteInputTokens` to `input_tokens` in Bedrock transformation
2. Update test assertions to reflect correct behavior
3. Add regression test for prompt caching cost calculation
4. Fix typo in Anthropic transformation where `cache_creation_tokens` was
   incorrectly set to `cache_read_input_tokens`

**Testing:**
- All existing Bedrock transformation tests pass
- New test validates correct cost calculation with prompt caching
- Verified costs are non-negative and accurate
 2025-10-07 20:28:46 +02:00