Commit Graph

30881 Commits

Author SHA1 Message Date
yuneng-jiang
cb1809987d Migrate some queries to use react query, tests pending 2025-11-25 21:00:41 -08:00
ripepersimmon
be970735de
feat: Add gemini-3-pro-image-preview model support for imageSize parameters (#17019)
- Add model identifier to FLASH_IMAGE_PREVIEW_MODEL_IDENTIFIERS
- Add imageSize parameter support (1K, 2K, 4K) with GeminiImageSize type
- Add tests for imageSize parameter transformation
- Update documentation with new model
2025-11-25 19:38:29 -08:00
Carlo Alberto Ferraris
a727f71b19
Optimize date filtering for spend logs queries (#17073)
This should allow postgres to perform a more efficient index scan instead of a sequential table scan.

These two queries consistently show up in the longest-running ones in our instance, and are a major latency source for the usage page on the admin UI.
2025-11-25 19:36:25 -08:00
Igal Boxerman
e6e1e8fca4
feat(pillar): add automatic LiteLLM context headers (#17076)
- Automatically pass LiteLLM virtual key context as X-LiteLLM-* headers
- Includes key_alias, user_id, team_id, org_id, and user_email
- No configuration required - always enabled for application/user tracking
- Excludes sensitive data (metadata, API tokens) for security
- Add comprehensive tests (30 tests, all passing)
- Update documentation with header details
2025-11-25 19:35:39 -08:00
Alexsander Hamir
7c09187daf
downgrade grpcio (#17090) 2025-11-25 19:33:24 -08:00
yuneng-jiang
7227747a6f
Improve Wording for Config Models in Model Table (#17100) 2025-11-25 19:25:23 -08:00
Carlo Alberto Ferraris
b50fcc4b56
vertex ai: use the correct domain for the global location when counting tokens (#17116) 2025-11-25 19:22:20 -08:00
Sameer Kankute
cd65a84abd
Merge pull request #16844 from Chesars/fix/response-format-to-text-format-bridge-conversion
fix: Support response_format parameter in completion -> responses bridge
2025-11-26 08:51:09 +05:30
yuneng-jiang
577f40bc60 ProviderLogo component, test pending 2025-11-25 18:05:30 -08:00
Ishaan Jaff
5c192a23c3
[Feat] Add new RAG API on LiteLLM AI Gateway (#17109)
* init RAG api types

* add RAG endpoints

* init main.py for RAG ingest API

* init RecursiveCharacterTextSplitter

* add BaseRAGIngestion

* fix OpenAIRAGIngestion

* fix img handler

* init OpenAIRAGIngestion

* init BedrockRAGIngestion

* init BedrockRAGIngestion

* init rag tests

* init BedrockVectorStoreOptions

* implement BedrockRAGIngestion

* add BaseRAGAPI

* add endpoint for RAG ingest

* add ingest RAG endpoints

* add test doc

* add parse_rag_ingest_request

* update endpoints

* docs add docs for new RAG API

* fix qa check

* fix linting

* docs ficx

* docs

* add max depth checks

* docs anthropic
2025-11-25 17:54:29 -08:00
yuneng-jiang
5ec3f19a53 Make model select required for team, add checks for all-proxy-models 2025-11-25 16:57:38 -08:00
Otavio Brito
6e5c7c0008
fix transcription exception handling - /audio/transcriptions (#16791)
* fix transcription exception handling

* reraise the exception
2025-11-25 16:41:35 -08:00
Krrish Dholakia
5cb5c2a7b7 docs: more doc cleanup 2025-11-25 16:04:27 -08:00
Krrish Dholakia
8ee6812edf docs: cleanup launch post 2025-11-25 15:58:51 -08:00
yuneng-jiang
3da9974a87 Tests 2025-11-25 15:54:55 -08:00
Krrish Dholakia
70a1325847 docs: more doc cleanup 2025-11-25 15:01:22 -08:00
Kerem Turgutlu
8637d74e17
include server_tool_use in streaming usage (#16826)
* include server_tool_use in streaming usage

* add test
2025-11-25 14:50:17 -08:00
Sam Chou
c0288d81aa
Fix bedrock claude opus 4.5 inference profile - only global currently (#17101) 2025-11-25 14:49:12 -08:00
yuneng-jiang
db587926a4 Sorting changes, pending tests and loading state 2025-11-25 14:46:46 -08:00
Krrish Dholakia
f3d5775920 fix: fix doc load issue 2025-11-25 14:40:26 -08:00
YutaSaito
52f1bf1a80
fix: missing await (#17103) 2025-11-25 14:33:38 -08:00
AlexsanderHamir
b3348c665a fix: use cached import helper for Logging in ahealth_check to preserve lazy loading
- Use get_litellm_logging_class() from cached_imports instead of direct import
- Preserves lazy loading benefit (only loads when function is called)
- Follows existing codebase pattern for cached imports
- Fixes NameError: name 'Logging' is not defined in test_ahealth_check_ocr
2025-11-25 14:25:56 -08:00
AlexsanderHamir
48810e6bcb refactor: improve MCP TextContent serialization to follow existing patterns
- Remove risky __dict__ fallback for non-BaseModel objects
- Only convert BaseModel objects to dicts using model_dump() (consistent with line 4745-4746)
- Keep other objects unchanged to maintain backward compatibility
- Follows existing codebase patterns for Pydantic model serialization
2025-11-25 13:14:25 -08:00
AlexsanderHamir
ab877d9551 fix: convert MCP TextContent objects to JSON-serializable format in logging
- Convert Pydantic BaseModel objects (TextContent, ImageContent, etc.) to dicts in get_final_response_obj
- Fixes TypeError: Object of type TextContent is not JSON serializable
- Resolves test failures in test_mcp_tool_call_hook and test_mcp_cost_tracking
2025-11-25 13:07:09 -08:00
AlexsanderHamir
2ef5a41a24 fix: resolve type checking errors in vertex_ai and mcp_server_manager
- Fix type incompatibility in vertex_ai/videos/transformation.py by casting litellm_params to Dict[str, Any]
- Add await to async add_update_server call in mcp_server_manager.py
- Resolves 4 type checking errors across 2 files
2025-11-25 12:48:24 -08:00
AlexsanderHamir
7f991fe897 refactor: clean up lazy loading system and remove type ignore comments
- Remove excessive comments and simplify documentation
- Remove @overload decorators (type stubs are sufficient)
- Remove Logging type stub to avoid redefinition errors
- Keep only essential type stubs for cost_per_token and completion_cost
- Fixes type checking errors without using type: ignore comments
2025-11-25 12:37:54 -08:00
Ishaan Jaff
be712908a3
[Feat] Add OpenAI compatible bedrock imported models. - qwen etc (#17097)
* test_bedrock_openai_imported_model

* AmazonBedrockOpenAIConfig

* add openai route for bedrock

* docs fix

* fix code qa check
2025-11-25 12:20:39 -08:00
AlexsanderHamir
36f8c9463f fix: resolve type checking errors for lazy-loaded functions in budget_manager
- Add type stubs and @overload decorators for cost_per_token and completion_cost
- Refactor lazy loading system with centralized registry for better maintainability
- Add comprehensive documentation for adding new lazy-loaded functions
- Fixes 'Any? not callable' errors at lines 111, 139, and 146 in budget_manager.py
2025-11-25 12:13:19 -08:00
yuneng-jiang
44cde2e48f Disable edit, delete, info, for dynamically generated spend tags 2025-11-25 12:03:01 -08:00
Krrish Dholakia
db2c8e3631 docs: initial doc cleanup 2025-11-25 11:57:51 -08:00
Sameer Kankute
67622fb040
Add day 0 support for anthropic new feat (#17091)
* Added tool search support for anthropic

* Add programtic tool calling support

* Add tool use input examples support

* Add anthropic effort param support

* Add anthropic effort param support

* Add blog for new features

* fix mypy and lint errors

* fix mypy and lint errors

* fix mypy and lint errors

* fix mypy and lint errors

* Add better handling

* Add better handling
2025-11-25 11:28:47 -08:00
abi_jey
aec0ab777b feat: add GA protocol as litellm_params for realtime api on azure provider 2025-11-25 19:18:41 +00:00
yuneng-jiang
359025554e bump: version 0.4.6 → 0.4.7 2025-11-25 11:05:49 -08:00
yuneng-jiang
42fa19a152 Revert "Test prisma changes for Gemini tests"
This reverts commit 031bb3c5d9.
2025-11-25 11:02:45 -08:00
Sameer Kankute
3249f6dd2d
Merge pull request #17070 from BerriAI/litellm_add_vertex_ai_image_support
Add vertex ai image gen support for both gemini and imagen models
2025-11-26 00:04:03 +05:30
Sameer Kankute
83a9dcd2d2
Merge pull request #16886 from BerriAI/litellm_anthopic_azure_support
Added support for azure anthopic models via chat completion
2025-11-26 00:03:52 +05:30
Sameer Kankute
59b4b9a07c fix documentation of anthropic azure 2025-11-26 00:02:48 +05:30
Sameer Kankute
59bcf079fb
Merge pull request #17078 from BerriAI/litellm_add_search_logging
Add search API logging and cost tracking in LiteLLM Proxy
2025-11-25 23:59:41 +05:30
Sameer Kankute
2e50db81a5
Merge pull request #17071 from BerriAI/litellm_azure_gpt_5_reasoning
Fix `reasoning_effort="none"` not working on Azure for GPT-5.1
2025-11-25 23:59:25 +05:30
Krish Dholakia
00e17c81a1
Add enforce user param functionality (#17088)
* feat: Add reject_metadata_tags to proxy config

Co-authored-by: krrishdholakia <krrishdholakia@gmail.com>

* Refactor: Rename reject_metadata_tags to reject_clientside_metadata_tags

Co-authored-by: krrishdholakia <krrishdholakia@gmail.com>

---------

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
2025-11-25 09:36:24 -08:00
Sameer Kankute
1c612288bc fix lint errors 2025-11-25 20:20:09 +05:30
Sameer Kankute
255d1bc239 fix lint errors 2025-11-25 20:20:09 +05:30
Sameer Kankute
e0396e5fa7
Merge pull request #17082 from BerriAI/main
merge main
2025-11-25 18:49:52 +05:30
Sameer Kankute
e2f2ccd913 Add tests related messages api 2025-11-25 18:45:51 +05:30
Sameer Kankute
dd4c8ecbef Add v1/messages support for azure anthropic models 2025-11-25 18:36:39 +05:30
Sameer Kankute
afe540e88d Fix auth issue 2025-11-25 18:26:25 +05:30
naaa760
2cf86e8ef8 new mocked 2025-11-25 17:28:02 +05:30
naaa760
a24b43cbdf only added the helper that converts 2025-11-25 17:27:39 +05:30
naaa760
2cc2f67a76 added the minimal GET logic 2025-11-25 17:27:12 +05:30
naaa760
c54986c3c9 list path now routes to Vertex 2025-11-25 17:26:53 +05:30