Ishaan Jaffer
859474dd44
fix patch
2025-12-20 14:00:42 +05:30
Ishaan Jaffer
13fa573d61
code QA fixes
2025-12-20 13:56:04 +05:30
Ishaan Jaffer
c1b116c5f5
fix
2025-12-20 13:54:05 +05:30
Cesar Garcia
138b415e81
fix(responses-api): use list format with input_text for tool results ( #18257 )
...
The Responses API expects tool results to use input_text/input_image types,
not output_text. This fix ensures consistent list format for all tool results:
- String content → [{"type": "input_text", "text": "..."}]
- Image content → [{"type": "input_image", "image_url": "..."}]
This resolves the conflict between tests that expected different formats
and aligns with OpenAI's Responses API requirements.
Fixes the regression introduced in #18226 .
2025-12-20 13:46:14 +05:30
Ishaan Jaffer
f3587ddfb0
linting fix
2025-12-20 13:44:28 +05:30
Krrish Dholakia
0ff13453ea
fix(types/utils.py): add order param to litellm_params
...
Fixes https://github.com/BerriAI/litellm/issues/18259
2025-12-20 13:30:11 +05:30
Krrish Dholakia
a3b6c6c0a7
fix: fix doc
2025-12-20 13:28:58 +05:30
Lucas Rothman
b981fddfaa
fix(gemini): properly catch context window exceeded errors
...
Fixes #18282
This PR fixes two issues with Gemini context window error handling:
1. **Pattern matching for Gemini 2.0 Flash**: The previous pattern
'input token count exceeds the maximum number of tokens allowed'
doesn't match Gemini 2.0 Flash errors which include dynamic token
counts like '(2800010)' in the message. Split into shorter patterns
that work with both formats.
2. **Add context window check to Gemini block**: The
is_error_str_context_window_exceeded() check was only called for
OpenAI-compatible providers, not for Gemini/Vertex AI. Added the
check to the Gemini-specific error handling block.
Test cases added for both Gemini 2.0 Flash and 2.5/3 error formats.
2025-12-19 23:20:04 -08:00
Eric84626
7a122ac350
Merge pull request #5 from Eric84626/feature/fix_mcp_closed_resource_error
...
fix: upgraded mcp sdk depency version for fixing ClosedResourceError
2025-12-20 14:29:40 +08:00
Eric84626
36a369a747
fix: upgraded mcp sdk depency version for fixing ClosedResourceError
2025-12-20 14:28:52 +08:00
Eric84626
ce3e598fa7
Merge pull request #4 from Eric84626/feature/fix_mcp_401_error_startup_proxyserver
...
fix: removed initialize the tool name to MCP server name mapping(oaut…
2025-12-20 14:14:52 +08:00
Eric84626
0306f02e74
fix: removed initialize the tool name to MCP server name mapping(oauth2) on startup for avoiding 401 error
2025-12-20 14:13:18 +08:00
Eric84626
b4d61f4f60
Merge pull request #3 from Eric84626/feature/add_scopes_supported
...
fix: added RFC RECOMMENDED property(scopes_supported) to protected re…
2025-12-20 13:28:42 +08:00
Eric84626
684fba42ea
fix: added RFC RECOMMENDED property(scopes_supported) to protected resource and authorization server metadata
2025-12-20 13:22:29 +08:00
Eric84626
c1954952c6
Merge pull request #2 from Eric84626/feature/add_grant_type_for_mcp_register_api
...
fix: added additional grant type into oauth_authorization_server resp…
2025-12-20 12:17:52 +08:00
Eric84626
c5f5867877
Merge pull request #1 from Eric84626/feature/fix_metadata_url_auth_servers_url_with_rootpath
...
fix: fixed the issue of handling root paths when processing Discovery…
2025-12-20 12:16:35 +08:00
Eric84626
3a2ab6b0d1
fix: added additional grant type into oauth_authorization_server response for fixing mcp auth register bad request issue
2025-12-20 11:55:12 +08:00
Eric84626
7ddab06bed
fix: fixed the issue of handling root paths when processing Discovery protected resource metadata and authorization server metadata URLs.
2025-12-20 11:35:20 +08:00
yuneng-jiang
ffcac2eebc
Allow deleting key expiry
2025-12-19 18:04:04 -08:00
Eric84626
96f7529f11
Merge branch 'BerriAI:main' into main
2025-12-20 09:42:41 +08:00
yuneng-jiang
4ca211a6f7
Merge pull request #18275 from BerriAI/litellm_ui_resolve_team_id
...
[Feature] UI - Resolve Team ID to Team Alias in Usage Page
2025-12-19 16:52:00 -08:00
yuneng-jiang
87e915e4e2
Resolve team alias in usage page
2025-12-19 16:43:52 -08:00
yuneng-jiang
bf72427b80
Merge pull request #18274 from BerriAI/litellm_ui_team_vs_fix
...
[Fix] UI - Persist Vector Store Settings in Team Update
2025-12-19 16:02:27 -08:00
YutaSaito
9ccb9f45e8
Merge pull request #18273 from BerriAI/revert-18268-litellm_fix_datadog-agent-connection-with-traces
...
Revert "ensure datadog llm obs ignores dd base url override"
2025-12-20 08:57:17 +09:00
YutaSaito
2bbfa498e2
Revert "ensure datadog llm obs ignores dd base url override"
2025-12-20 08:56:49 +09:00
yuneng-jiang
8a8b4de37f
Persist vector stores in update team settings
2025-12-19 15:54:13 -08:00
yuneng-jiang
a6347aa3a6
Merge pull request #18269 from BerriAI/litellm_ui_wildcard_health_checks
...
[Feature] UI - Add Health Check Model for Wildcard Model in UI
2025-12-19 15:28:07 -08:00
yuneng-jiang
40b823af87
Add Health Check Model for Wildcard in UI
2025-12-19 15:13:06 -08:00
YutaSaito
28cda220c7
Merge pull request #18268 from BerriAI/litellm_fix_datadog-agent-connection-with-traces
...
ensure datadog llm obs ignores dd base url override
2025-12-20 08:12:10 +09:00
Yuta Saito
72f424c719
ensure datadog llm obs ignores dd base url override
2025-12-20 07:44:40 +09:00
Alexsander Hamir
d9181c188e
[Refactor] - Lazy load 41 configuration classes ( #18267 )
2025-12-19 14:19:22 -08:00
Alexsander Hamir
5b1fda02fb
Add infrastructure recommendations to benchmarks documentation ( #18264 )
...
Added concise PostgreSQL and Redis specifications based on benchmark results and industry standards for API gateway deployments. Includes tiered recommendations for different RPS workloads, configuration best practices, and scaling guidelines.
2025-12-19 13:44:36 -08:00
yuneng-jiang
74842de78e
Adding backend
2025-12-19 13:04:59 -08:00
yuneng-jiang
e92db2d718
Improve e2e flow for cloudzero
2025-12-19 13:04:42 -08:00
Alexsander Hamir
f3b97356fb
Add lazy loading for GaladrielChatConfig to reduce import memory overhead ( #18260 )
...
Implements lazy loading pattern for GaladrielChatConfig following the existing approach used for other LLM config classes. This defers the import until the config is actually accessed, reducing memory usage during module initialization.
2025-12-19 13:00:57 -08:00
yuneng-jiang
026c2ad693
Merge pull request #18167 from BerriAI/litellm_vector_store_config
...
[Feature] Auto Resolve Vector Store Embedding Model Config
2025-12-19 11:39:17 -08:00
yuneng-jiang
cacaf54b95
Merge pull request #18170 from BerriAI/litellm_ui_cloudzero_delete
...
[Feature] UI - Cloudzero Delete Settings
2025-12-19 11:38:28 -08:00
yuneng-jiang
b1b488982d
Merge pull request #18168 from BerriAI/litellm_cloudzero_delete
...
[Feature] Delete Cloudzero Settings Route
2025-12-19 11:38:10 -08:00
yuneng-jiang
02355f602c
Merge remote-tracking branch 'origin' into litellm_sso_role_mapping
2025-12-19 11:03:30 -08:00
yuneng-jiang
313a613a13
Adding tests
2025-12-19 10:56:11 -08:00
Sameer Kankute
3761f38e43
Merge pull request #18254 from BerriAI/litellm_add_stability_model_edit_1
...
Add support for stability model and bedrock stability model
2025-12-20 00:23:15 +05:30
Sameer Kankute
2962bcac63
Merge pull request #18256 from BerriAI/litellm_rever_proxy
...
revert proxy_server_config.py
2025-12-20 00:20:47 +05:30
Sameer Kankute
d3fb7528d5
revert proxy_server_config.py
2025-12-20 00:20:20 +05:30
Sameer Kankute
b3d17951cd
revert proxy_server_config.py
2025-12-20 00:19:40 +05:30
Sameer Kankute
e430e317f5
revert proxy_server_config.py
2025-12-20 00:18:03 +05:30
Sameer Kankute
f05e600969
Merge pull request #18236 from BerriAI/litellm_add_background_responses_cost_tracking
...
Add cost tracking for responses api in background mode
2025-12-20 00:14:09 +05:30
Sameer Kankute
d29f4cab59
Merge pull request #17971 from BerriAI/litellm_ocr_deepseek
...
Add support for ocr for vertex ai deepseek model
2025-12-20 00:13:57 +05:30
Sameer Kankute
3d56d3d49b
vertex ai deepseek doesn't support pdf urls, so skip them
2025-12-19 23:53:17 +05:30
Sameer Kankute
b0a219b718
fix lint error
2025-12-19 23:45:37 +05:30
Sameer Kankute
8efd4653ce
Fix litellm_mapped_tests_llms
2025-12-19 23:44:14 +05:30
