Commit Graph

30881 Commits

Author SHA1 Message Date
Alexsander Hamir
29a31e17dd
[Doc] Perf: Last week improvement (#15193)
* doc: perf update

* fix: mixed up changes

* fix: add gist
2025-10-04 10:43:13 -07:00
Ishaan Jaffer
ee36c30217 fix LF tests 2025-10-04 10:39:17 -07:00
Ishaan Jaffer
00f44861ea fix: gooogle GenAI route tests 2025-10-04 10:18:25 -07:00
Ishaan Jaffer
df9a19bc9d fix OPENAI_EMBEDDING_PARAMS 2025-10-04 10:11:21 -07:00
Ishaan Jaffer
1655a9aea8 fix lf tests 2025-10-04 10:02:56 -07:00
Ishaan Jaffer
eb417ed774 fix lf logging 2025-10-04 10:00:10 -07:00
Ishaan Jaffer
27fbfbe259 fix: include_subpath 2025-10-04 09:47:16 -07:00
Ishaan Jaffer
212a2e13f3 OTEL fix spans 2025-10-04 09:22:57 -07:00
Ishaan Jaffer
a040e0154a add openai 2025-10-04 09:19:00 -07:00
Ishaan Jaffer
4400a6c189 test bedrock guardrails 2025-10-04 09:18:26 -07:00
Alexsander Hamir
5d22229d35
[Fix] Cache - Avoiding expensive operations when cache isn't available (#15182)
* Optimize cache performance by avoiding expensive operations when caching is disabled

- Moved cache availability checks before expensive operations to improve performance for non-cached requests
- Updated client code to handle None responses from caching handler

* clean hot path

* Fix TypeError with isinstance check for CustomStreamWrapper in caching

Fixed `TypeError: typing.Any cannot be used with isinstance()` that was
occurring in the caching handler when checking cached streaming responses.

The issue was caused by CustomStreamWrapper being aliased to `typing.Any`
at runtime through the TYPE_CHECKING conditional import pattern. When the
code attempted to use isinstance(cached_result, CustomStreamWrapper) at
lines 222 and 338, it failed because Python's isinstance() cannot be used
with typing.Any.

Solution: Import CustomStreamWrapper at runtime separately from the
TYPE_CHECKING block, while keeping a type alias for static type checking.
This allows isinstance checks to work properly while maintaining type hints.

* fix: remove unnecessary type checking
2025-10-04 09:10:37 -07:00
Ishaan Jaffer
d50fbcdc00 fix: include_cost_in_streaming_usage 2025-10-04 09:05:02 -07:00
Ishaan Jaffer
89934f062c fix ruff check 2025-10-04 08:59:57 -07:00
Ishaan Jaffer
3d981680b0 fix ServerToolUse mypy lint error 2025-10-04 08:59:15 -07:00
Ishaan Jaffer
8e7c593f51 fix: transform_rerank_response 2025-10-04 08:55:48 -07:00
Krrish Dholakia
f5cda40bc6 feat: prisma migration 2025-10-04 08:53:08 -07:00
Ishaan Jaffer
67b0c874a3 ci/cd run again 2025-10-04 08:51:21 -07:00
Krish Dholakia
ad2270fe13
Merge pull request #14764 from daily-kim/litellm_fix_bearer_capitalization
Fix: Authorization header to use correct "Bearer" capitalization
2025-10-03 22:02:46 -07:00
Krish Dholakia
7d41f02d02
Merge pull request #14813 from shagunb-acn/bugfix-14404-image-gen-azure-managed-identity
#14404 BugFix - Add support for Azure AD token-based authorization in…
2025-10-03 21:53:59 -07:00
Krish Dholakia
c83a3ac9b4
Merge pull request #14939 from Toy-97/patch-1
update: DeepInfra model data refresh [2025-09-26]
2025-10-03 21:48:31 -07:00
Krish Dholakia
206f67a11e
Merge pull request #15183 from uc4w6c/fix/test_mcp_server
test: fix test_mcp_server.py
2025-10-03 21:44:52 -07:00
YutaSaito
81a8766b84
feat: add JP Cross-Region Inference (#15188) 2025-10-03 21:20:04 -07:00
Kowyo
981dd88792 Merge branch 'kowyo/fix-ollama-think' of github.com:kowyo/litellm into kowyo/fix-ollama-think 2025-10-04 02:50:28 +00:00
Kowyo
bb3829b697 fix: update 'think' parameter assignment to use provided value in transformation.py 2025-10-04 02:50:00 +00:00
Kowyo
ef90bb8680
Merge branch 'BerriAI:main' into kowyo/fix-ollama-think 2025-10-04 10:35:20 +08:00
Krrish Dholakia
39cac8e58a build(ui/): show updated schema.prisma 2025-10-03 19:07:50 -07:00
Krrish Dholakia
aa1b690494 feat(ui/): allow user to edit allowed tools for an mcp server 2025-10-03 19:05:56 -07:00
=
aca8ae7962 fixes issue where empty premium fields were blocking key edit 2025-10-03 18:57:50 -07:00
Krrish Dholakia
6691d0bc85 feat(proxy/): add edit / view for extra headers on MCP servers 2025-10-03 18:57:27 -07:00
Krrish Dholakia
c00cedcf26 feat(ui/): display the MCP extra headers on the UI
allow user to see what extra headers they set
2025-10-03 18:43:41 -07:00
Krrish Dholakia
562e7a4e03 feat(proxy/_types.py): add new extra_headers param support to API request
allows admin to specify which headers they want to forward to the backend MCP server
2025-10-03 18:29:37 -07:00
Krrish Dholakia
274c5e963d feat(ui/): allow user to forward additional headers to the MCP server from the UI 2025-10-03 18:22:11 -07:00
Krrish Dholakia
950210790b feat(create_mcp_server.tsx): initial ui component for adding extra headers, allowed tools, disallowed tools to litellm ui
Closes LIT-1183
2025-10-03 18:12:31 -07:00
Krrish Dholakia
72b5cf2b71 fix(schema.prisma): add new db columns 2025-10-03 17:57:25 -07:00
Krrish Dholakia
45f7f86480 feat(mcp_servers.tsx): show health of mcp servers on UI
allow user to know if mcp server is in invalid state
2025-10-03 17:42:06 -07:00
Krrish Dholakia
3816d3b36f test: add unit testing 2025-10-03 17:27:49 -07:00
Krrish Dholakia
6ccb20d69d fix(mcp_server_manager.py): mark mcp server as unhealthy and show reason for not being able to serve it on list servers on ui
make it easy for user to debug why server can't be added
2025-10-03 17:25:08 -07:00
Krrish Dholakia
c043ad6c2b fix: replace list in for set in
faster
2025-10-03 17:17:20 -07:00
Krrish Dholakia
69edc546c0 fix(mcp_server_manager.py): don't return an invalid server id on list servers
Prevents user from hitting 'server not found' error, even when they see it being listed on litellm ui
2025-10-03 17:15:15 -07:00
Krrish Dholakia
742038bb5e build: deploy/
push new prisma migration
2025-10-03 16:52:03 -07:00
Yuta Saito
359aaa947f test: fix test_mcp_server.py 2025-10-04 08:18:15 +09:00
Ishaan Jaff
10d6d72ae3
[Feat] VertexAI - Support googlemap grounding in vertex ai (#15179)
* add VertexToolName

* test_vertex_tool_params

* fix: working maps grounding

* test_gemini_google_maps_tool_simple

* test_vertex_ai_map_google_maps_tool_with_location

* fix  # noqa: PLR0915

* _extract_google_maps_retrieval_config

* fixes for linting

* docs: **Google Maps**
2025-10-03 16:07:51 -07:00
Ishaan Jaff
4415b195d1
Add "eu.anthropic.claude-sonnet-4-5-20250929-v1:0" in "model_prices_and_context_window.json" (#15181)
* feat: add eu.anthropic.claude-sonnet-4-5-20250929-v1:0

* fix: nvidia_nim_models
2025-10-03 15:52:18 -07:00
Krish Dholakia
3095f43e92
Merge pull request #15007 from TobiMayr/feature/add-max-requests-env-var
feature/add max requests env var
2025-10-03 15:30:15 -07:00
Krish Dholakia
c5de073732
Merge pull request #15072 from speglich/fix/oci-support
Fix OCI Generative AI  Integration when using Proxy
2025-10-03 15:29:21 -07:00
Krish Dholakia
343eeaf53f
Merge pull request #15102 from BerriAI/litellm_message_api_cost_tracking
(Feat) Add cost tracking for /v1/messages
2025-10-03 15:27:49 -07:00
Krish Dholakia
b41041fc49
Merge pull request #15146 from JVenberg/fix/session-cookie-infinite-logout-loop
[Fix] Session Token Cookie Infinite Logout Loop
2025-10-03 15:23:49 -07:00
Krish Dholakia
f05dc27551
Merge pull request #15160 from danielaskdd/fix-whitespace-handling
Fix(critical) : Preserve Whitespace Characters in Model Response Streams
2025-10-03 15:23:07 -07:00
rishiganesh2002
d36c8d6bbe
[Feat] MCP Gateway Fine-grained Tools Addition (#15153)
* feat: UI to add specific tools under creating MCP connection

* chore: pydantic + prisma changes

* feat: adding specific MCP tools now works

* fix: allowed tools filtering

* chore: filtered list to mcp server cost config

* chore: update Readme

* chore: refactor the filtering

* test: Added tests

When the allowed_tests is null, empty list or populated

* chore: resolve the proxy issue

* feat: updating MCP tool filtering
2025-10-03 10:16:29 -07:00
yangdx
8b95cd3ed2 Fix whitespace handling in _has_meaningful_content function
• Preserve newlines and spaces as content
• Remove .strip() call on strings
• Treat all non-empty strings as meaningful
• Update logic comment for clarity
• Fix edge case with whitespace-only text
2025-10-03 13:46:51 +08:00