tiennm99/litellm
1
0
Fork 0
mirror of https://github.com/tiennm99/litellm.git synced 2026-08-20 06:23:46 +00:00
Code Issues Packages Projects Releases Wiki Activity
30,881 Commits 2,072 Branches 1,425 Tags
12bc66aa5ba767f2af4daf4ae99d41af77ba5ede
Commit Graph
5648 Commits
Author SHA1 Message Date
YutaSaitoandGitHub 4b131a77a0 Merge pull request #18573 from BerriAI/litellm_fix_normalize-model-name 
fix: unify model names to provider-defined names
 2026-01-02 17:37:28 +09:00
Lukas de BoerandGitHub edc8413f1e Add Kubernetes ServiceAccount JWT authentication support (#18055) 
* Allow get_nested_value dot notation to support escaping for Kubernetes JWT Support

* Add support for team and org alias fields, add docs, tests

* Fix lint issue with max statements in handle jwt logic
 2026-01-02 14:02:31 +05:30
Yuta Saito 0e01d225ff fix: test_edit_mcp_server_redacts_credentials error 2026-01-02 17:13:31 +09:00
Yuta Saito 89ce8091bb fix: test 2026-01-02 16:32:33 +09:00
Sameer Kankute 57c373c5d8 Add validation for negative budget 2026-01-02 12:57:39 +05:30
Sameer Kankute dd96f91254 Add thought signature for non tool call requests 2026-01-02 12:31:31 +05:30
Yuta Saito 3ae38c2a6d fix: test_langfuse_logging_completion_with_bedrock_llm_response 2026-01-02 15:55:02 +09:00
Yuta Saito 35f9a75d55 feat: add UI support for configuring meta URLs 2026-01-02 15:07:37 +09:00
Sameer Kankute ca94990307 remove prompt caching headers as the support has been removed 2026-01-02 11:08:35 +05:30
Sameer KankuteandGitHub 7ba14a30ed Merge pull request #18500 from systec-dk/fix_anthropic 
fix authentication errors at messages API via azure_ai
 2026-01-02 09:22:12 +05:30
mangabits 4a1bc90e61 Add more tests 2026-01-01 19:08:34 -08:00
mangabits 78693bb9d0 Use already configured opentelemetry providers 
Users that instrument using opentelemetry-instrument can now setup exporters as per their environment.
 2026-01-01 19:06:55 -08:00
Yuta Saito f7241a7905 fix: unify model names to provider-defined names 2026-01-02 08:36:58 +09:00
Alexsander HamirandGitHub d2f2fe1be5 [Fix] CI/CD - litellm_mapped_tests_proxy (#18572) 2026-01-01 15:27:51 -08:00
Alexsander HamirandGitHub 3f6312c0ef [Fix] CI/CD - litellm_mapped_tests_llms (#18563) 2026-01-01 11:46:32 -08:00
Alexsander HamirandGitHub a7da4833da [Fix] CI/CD - check_code_and_doc_quality (#18560) 2026-01-01 11:18:59 -08:00
yuneng-jiangandGitHub 9be2173d5b Merge pull request #18557 from BerriAI/litellm_deactivate_other_ui_token 
[Feature] Expire Previous UI Session Tokens on Login
 2026-01-01 09:11:17 -08:00
yuneng-jiang 269bc5b26b Expire Previous UI Session Tokens 2025-12-31 19:26:59 -08:00
YutaSaitoandGitHub faf58dc351 Merge pull request #18553 from BerriAI/litellm_fix_ends_parent_span_in_otel 
fix: prevent LiteLLM from closing external OTEL spans
 2026-01-01 11:49:12 +09:00
Yuta Saito b343d15157 fix: prevent LiteLLM from closing external OTEL spans 2026-01-01 08:28:48 +09:00
xuan07t2andClaude Opus 4.5 f3f1acd338 fix(vertex_ai): separate Tool objects for each tool type per API spec 
Fix Vertex AI API error: "tools[0].tool_type: one_of 'tool_type' has more
than one initialized field"

The Vertex AI API requires each Tool object to contain exactly one type
of tool (e.g., FunctionDeclaration, GoogleSearch, CodeExecution).
Previously, all tool types were combined into a single Tool object,
causing INVALID_ARGUMENT errors when using multiple tools simultaneously.

This change creates separate Tool objects for each tool type:
- Function declarations in one Tool
- Google Search in its own Tool
- Code Execution in its own Tool
- etc.

Ref: https://cloud.google.com/vertex-ai/generative-ai/docs/reference/rest/v1beta1/Tool

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
 2025-12-31 08:48:35 +07:00
YutaSaitoandGitHub cc8e310736 Merge pull request #18530 from BerriAI/litellm_refactor_mcp_health_check 
feat: optimize MCP server listing by separating health checks
 2025-12-31 08:03:12 +09:00
Yuta Saito 7fdea85b5c feat: optimize MCP server listing by separating health checks 2025-12-31 06:55:47 +09:00
hamzaq453 4573ab326b refactor: remove to_safe_identifier mapping from OpenAPI MCP generator 2025-12-30 14:04:15 +05:00
yuneng-jiangandGitHub ff4e16b306 Merge pull request #18502 from BerriAI/litellm_key_list_user_object 
[Feature] Add Optional Query Param "expand" to /key/list
 2025-12-29 15:02:38 -08:00
YutaSaitoandGitHub b9e8b38cd3 Merge branch 'main' into fix/mcp-rest-auth-ssrf 2025-12-30 07:22:01 +09:00
yuneng-jiang de551889f3 Add optional field expand to /key/list 2025-12-29 14:20:04 -08:00
Daniel Krueger d7468dab7e fix authentication errors at messages API via azure_ai 
Use x-api-key instead of api-key.
This has been removed by commit 61e737e361 for unknown reason.
 2025-12-29 15:29:54 +01:00
YutaSaitoandGitHub a569369195 Merge pull request #18443 from BerriAI/litellm_fix_health_status 
fix health status
 2025-12-29 14:16:38 +09:00
Devaj e4c9b0bea2 fix(vertex_ai): convert image URLs to base64 for Vertex AI Anthropic 
Fixes #18430

  - Pass custom_llm_provider to anthropic_messages_pt instead of hardcoded 'anthropic'
  - Add check for vertex_ai provider to force base64 conversion for image URLs
  - Add tests to verify behavior for both Vertex AI and regular Anthropic
 2025-12-29 09:55:41 +05:30
hamzaq453 b33f1ec2b2 Fix: Remove exec() usage and handle invalid OpenAPI parameter names 
- Add to_safe_identifier() to convert any parameter name to valid Python identifier
- Refactor create_tool_function() to use closure with **kwargs instead of exec()
- Handle edge cases: hyphens, dots, leading digits, Python keywords, special chars
- Add comprehensive test suite covering all edge cases
- Fixes #18471: OpenAPI MCP server crashes on invalid parameter names
- Security: Eliminates arbitrary code execution risk from untrusted OpenAPI specs
 2025-12-28 19:28:11 +05:00
Yuta Saito 6888d34ea5 feat: Log actual executed event type in guardrail logging 2025-12-27 07:20:24 +09:00
Yuta Saito c25ceb5596 refactor: MCP health check 2025-12-26 16:55:51 +09:00
YutaSaitoandGitHub 78b5d40664 Merge pull request #18418 from BerriAI/litellm_fix_datadog-llm-observability-intermittent-failure 
fix: Datadog span kind fallback when parent_id missing
 2025-12-26 10:05:08 +09:00
Andrew BernatGitHubPlan42.ai
 		6daddbcdab add license endpoint (#18311) 
* Add license health endpoint

* Address review feedback for license endpoint

* Refactor license health endpoint helpers

* Remove unused timezone import from health_endpoints.py

---------

Co-authored-by: Plan42.ai <robot@plan42.ai>
 2025-12-25 11:11:17 +05:30
Ishaan JaffandGitHub 1123cfa928 [Feat] AI Gateway - Add support for Platform Fee / Margins (#18427) 
* init cost_margin_config

* feat: add cost margin

* init types

* LITELLM_SETTINGS_SAFE_DB_OVERRIDES

* feat _apply_cost_margin

* ui endpoint

* ui provider margins

* add margin

* refactored ui

* test cost margins

* refactored ui

* provider discounts

* add cost_breakdown to spendLogs

* add CostBreakdownViewer

* fix cost breakdown

* docs fix

* doc margins

* docs margins
 2025-12-25 11:07:27 +05:30
yuneng-jiang 5a20edf0fa Allow org admins to view org info 2025-12-24 11:43:36 -08:00
Sameer KankuteandGitHub 109def6077 Merge pull request #18413 from BerriAI/litellm_azure_ad_image_generation 
Add support for image generation via azure ad token
 2025-12-24 21:42:34 +05:30
Sameer KankuteandGitHub fd7c928204 Merge pull request #18410 from BerriAI/litellm_refactor_location_code 
feat(vertex-ai): add centralized get_vertex_base_url() helper for global location support
 2025-12-24 21:42:18 +05:30
Sameer Kankute d88bc13006 Add support for image generation via azure ad token 2025-12-24 15:15:20 +05:30
Sameer Kankute d02c580779 feat(vertex-ai): add centralized get_vertex_base_url() helper for global location support 
- Add get_vertex_base_url() helper function to handle regional vs global URLs
- Update _get_embedding_url() to support global location
- Update _get_vertex_url() chat, image_generation, and count_tokens modes
- Add comprehensive test suite with 38 tests covering all endpoint types
- Tests verify both regional and global URL construction
- Maintains 100% backward compatibility
 2025-12-24 13:51:58 +05:30
Sameer Kankute 4c4c9cb353 Add generate content in llm route 2025-12-24 12:16:24 +05:30
Yuta Saito 61e1e42e97 fix: Datadog span kind fallback when parent_id missing 2025-12-24 14:12:17 +09:00
yuneng-jiang 564b2b51cc Fix for dev env 2025-12-23 16:09:17 -08:00
yuneng-jiang 05dd247ff5 Fix UI disappearing for development instances 2025-12-23 15:24:07 -08:00
yuneng-jiangandGitHub b86147c84c Merge pull request #18397 from BerriAI/litellm_disable_admin_flag 
[Fix] Disable Admin UI Flag
 2025-12-23 15:20:24 -08:00
yuneng-jiangandGitHub 6bb5254c9b Revert "[Fix] UI - Disappears in Development Environments" 2025-12-23 15:08:07 -08:00
yuneng-jiang d9bcc33ba7 Fixing tests 2025-12-23 14:10:41 -08:00
yuneng-jiang 87337abfea Support Disable Admin UI 2025-12-23 13:52:05 -08:00
YutaSaitoandGitHub 55bfb24ef8 Merge pull request #18324 from BerriAI/litellm_feat_dynamic_env_propagation_for_stdio_MCP_server 
feat: support MCP stdio header env overrides
 2025-12-24 06:29:53 +09:00