Sameer Kankute
8ab4f8cfe6
fix lint error
2025-12-19 23:40:23 +05:30
Alexsander Hamir
24edbccf5c
[Fix] Security - Remove example API keys with high entropy ( #18255 )
2025-12-19 10:09:50 -08:00
Sameer Kankute
e39376c54b
fix lint error
2025-12-19 23:39:37 +05:30
Sameer Kankute
06d08b52d9
Merge pull request #18237 from BerriAI/litellm_gemini_response_format_fix
...
Fix: properties: should be non-empty for OBJECT type
2025-12-19 23:34:05 +05:30
Sameer Kankute
f062dbdca6
Merge pull request #18227 from BerriAI/litellm_gemini3_doc_update
...
Add version requirement in gemini 3 flash blog
2025-12-19 23:30:20 +05:30
Sameer Kankute
86596de193
add vertex ai location note
2025-12-19 23:29:38 +05:30
Sameer Kankute
9edde2ffdb
Merge pull request #18239 from BerriAI/litellm_fix_sdk_import_error
...
fix:No module named 'fastapi'
2025-12-19 23:27:23 +05:30
Sameer Kankute
4dedb128b7
fix code qa check
2025-12-19 23:26:05 +05:30
Sameer Kankute
98382bdb93
Fix mypy errors
2025-12-19 23:21:20 +05:30
Sameer Kankute
1f566f6720
fix: lint erros
2025-12-19 23:16:27 +05:30
yuneng-jiang
bf76e66d2c
Working SSO Mapping new user, overrides default user settings
2025-12-19 09:42:59 -08:00
Sameer Kankute
248eb4ea49
Minor bug fixes: mask list
2025-12-19 23:04:53 +05:30
Sameer Kankute
5feb6d5388
Add support for stability speicifc params in request
2025-12-19 23:03:17 +05:30
Sameer Kankute
b849f51e58
Add support for stability model in image edit
2025-12-19 23:03:17 +05:30
Sameer Kankute
b09ea08424
Add stability models for image edit
2025-12-19 23:03:17 +05:30
yuneng-jiang
e1d25670cd
Merge pull request #18214 from BerriAI/litellm_key_management_fix
...
[Fix] Key Delete and Regenerate Permissions Fix
2025-12-19 08:48:21 -08:00
Ishaan Jaffer
388535003e
docs fix
2025-12-19 20:06:56 +05:30
Anil Kodali
afba676b2e
Add Amazon Nova to sidebar and under supported models in README ( #18220 )
2025-12-19 19:07:34 +05:30
Ishaan Jaff
deb8d16967
[Feat] Add new Rag Search API / Query API with rerankers ( #18217 )
...
* init RAGQueryRequest
* init RAGQuery
* fix query
* fix _execute_query_pipeline
* TestRAGOpenAI
2025-12-19 19:05:07 +05:30
Ishaan Jaffer
b0db9d6bb7
code qa check
2025-12-19 19:00:44 +05:30
Ishaan Jaff
09b5ee9428
[Feat] Unified Skills API - works across Anthropic, Vertex, Azure, Bedrock ( #18232 )
...
* init LiteLLM_SkillsTable
* init LiteLLMSkillsTransformationHandler
* init LiteLLMSkillsTransformationHandler
* init skills
* init SkillsInjectionHook
* init litellm skills handler
* _fetch_skill_from_db
* LiteLLMSkillsTransformationHandler
* add _prisma_skill_to_litellm
* use SkillPromptInjectionHandler
* refactor skills handler
* add slack GIF
* test_skill_code_execution_via_deployment_hook
* init readme
* test_get_skill_sdk
* test skills in DB
* fix code qa
* init with containers param
* SkillsInjectionHook
* fix type
* convert_skill_to_anthropic_tool, get_litellm_code_execution_tool_anthropic
* fix messages skills handler
* test_slack_gif_skill_creates_gif
* init SkillsInjectionHook
* clean async_pre_call_hook
* fixes
* working unified skills API! yeet
* fix mypy linting
* LiteLLM_SkillsTable
* llm-sandbox==0.3.31
* fix
2025-12-19 18:55:59 +05:30
Sameer Kankute
568fa73c5d
fix:No module named 'fastapi'
2025-12-19 15:20:01 +05:30
Sameer Kankute
a0538e3d54
Fix: properties: should be non-empty for OBJECT type
2025-12-19 14:28:40 +05:30
YutaSaito
3c380086b6
Merge pull request #18234 from BerriAI/litellm_fix_not_working_failure_logging
...
fix: not working log_failure_event in langfuse
2025-12-19 17:32:05 +09:00
Sameer Kankute
84e4fc3fab
Merge pull request #18226 from Chesars/fix/responses-api-tool-calls-transformation
...
fix(responses-api): fix tool calls transformation in completion bridge
2025-12-19 13:41:55 +05:30
Sameer Kankute
7d0f41f437
Add cost tracking for responses api in background mode
2025-12-19 13:35:48 +05:30
YutaSaito
369111d186
Merge pull request #18235 from uc4w6c/doc/improve_secret_manager
...
doc: improve secret manager settings
2025-12-19 17:03:29 +09:00
Yuta Saito
20bbfdd7cb
fix: not working log_failure_event in langfuse
2025-12-19 16:54:57 +09:00
yuneng-jiang
887aa62ac7
Ruff check
2025-12-18 21:09:11 -08:00
Sameer Kankute
9e3acdb815
Add version requirement in gemini 3 flash blog
2025-12-19 10:38:06 +05:30
Sameer Kankute
d7647f5b86
Add version requirement in gemini 3 flash blog
2025-12-19 10:17:58 +05:30
Sameer Kankute
d1d008fb7e
Merge pull request #18194 from BerriAI/litellm_fix_cli_bugs
...
Fix: Claude code responses api bridge errors
2025-12-19 08:38:55 +05:30
Sameer Kankute
1fd4f1ba52
Merge pull request #18109 from BerriAI/litellm_fix_custom_gaurdrail_fix
...
Fix guardrails for passthrough endpoint
2025-12-19 08:36:58 +05:30
Chesars
5a1cc51d0d
docs: add function calling example for Responses API
2025-12-18 22:07:04 -03:00
Chesars
bd36b1261f
test: add tests for tool calls transformation fixes
...
- test_tool_message_output_is_string_not_list: verifies function_call_output.output is a string
- test_multiple_tool_calls_in_single_choice: verifies multiple tool calls are grouped in one choice
2025-12-18 22:01:29 -03:00
Chesars
c60b2dd987
fix(responses-api): fix tool calls transformation in completion bridge
...
Fixes two bugs in the openai/responses/... completion bridge:
1. function_call_output.output must be a string, not a list
- When sending tool results back to the model, the content was being
transformed to [{type: "output_text", text: "..."}] instead of a plain string
- This caused OpenAI to reject with "Invalid value: 'output_text'"
2. Multiple tool calls must be in a single choice, not separate choices
- When the model returned multiple tool calls, each was put in its own
Choice with index 0, 1, 2... instead of all together in one Choice
- This broke the standard Chat Completions API format where all tool_calls
belong in a single message
Fixes #18201
2025-12-18 21:59:30 -03:00
yuneng-jiang
085d07db46
Fixing tests
2025-12-18 16:12:06 -08:00
Alexsander Hamir
0c48826cdc
Revert "[Fix] CI/CD - local_testing & mapped tests ( #18222 )" ( #18223 )
...
This reverts commit dc7f500c47
2025-12-18 14:46:09 -08:00
Alexsander Hamir
dc7f500c47
[Fix] CI/CD - local_testing & mapped tests ( #18222 )
2025-12-18 14:34:48 -08:00
Yuta Saito
854d5a8d66
doc: improve secret manager settings
2025-12-19 07:06:26 +09:00
yuneng-jiang
41732696c6
replicate delete checks for regenerate
2025-12-18 13:25:11 -08:00
Alexsander Hamir
5230e97448
[Fix] CI/CD - Add missing documentation for ENVs & Clean Functionality ( #18211 )
2025-12-18 13:16:15 -08:00
Chesars
e08b4767e9
fix: case-insensitive model cost map lookup
...
Users were getting "does not support parameters: ['tools']" errors when
using lowercase model names (e.g., "qwen/qwen3-next-80b-a3b-thinking")
because the model cost map has mixed-case keys and the lookup was
case-sensitive.
Added _get_model_cost_key() helper that tries exact match first (O(1)),
then falls back to case-insensitive search if not found.
2025-12-18 17:25:43 -03:00
Ishaan Jaff
5bffa30cb9
Update README.md
2025-12-19 01:35:48 +05:30
Ishaan Jaff
06d688abc9
Update README.md
2025-12-19 01:35:08 +05:30
Ishaan Jaff
36f28dbbb7
[Readme] fixes ( #18206 )
...
* v1
* fix
* docs fix
* Update README.md
* docs fix
* docs fix
* docs fix
* docs fix
* docs fix
* docs fix
* docs
* docs
2025-12-19 01:30:48 +05:30
Will Chen
3f686521ed
Fix Gemini 3 Flash pricing for Vertex AI ( #18202 )
2025-12-19 01:13:04 +05:30
yuneng-jiang
0a1fb204cd
Tests for /key/delete
2025-12-18 11:40:41 -08:00
yuneng-jiang
7a1e180573
Working /key/delete permission changes
2025-12-18 11:32:58 -08:00
yuneng-jiang
e0ef4fa8f2
Base commit
2025-12-18 11:31:26 -08:00
