Commit graph

26892 commits

Author SHA1 Message Date
Andrew Bernat
d89990e0c5
Add license metadata to health/readiness endpoint. (#15997)
* health: expose license metadata (available & expiration) in /health/readiness endpoint

* test: add health readiness license metadata coverage

* test: ensure /health/readiness response includes license metadata

* chore: remove standalone license metadata test as requested; existing test covers codepath

---------

Co-authored-by: Plan42.ai <robot@plan42.ai>
2025-10-28 19:21:54 -07:00
Ishaan Jaffer
a5b725917c fix merge 2025-10-28 19:20:07 -07:00
Ishaan Jaffer
f28e6fcbdd ui new build 2025-10-28 19:20:07 -07:00
Daniele Scasciafratte
36f0ee6ff9
Remove unnecessary model variable assignment (#16008)
* Remove unnecessary model variable assignment

Remove redundant assignment of model variable.

* Remove redundant model assignment in image generation
2025-10-28 19:19:25 -07:00
Ishaan Jaffer
33371d18f4 test fix claude-sonnet-4-5-20250929 2025-10-28 19:05:13 -07:00
Rodolfo Nobrega de Resende
29f0ed223a
fix: Support text.format parameter in Responses API for providers without native ResponsesAPIConfig (#16023)
Fixes #15995

When using the Responses API with providers that don't have a native
ResponsesAPIConfig implementation (like Gemini, Anthropic, Cohere, etc.),
the text.format parameter was being ignored. This happened because these
providers fall back to using the LiteLLMCompletionTransformationHandler,
which converts Responses API requests to Chat Completion API requests.

Changes:
- Added 'text' to the list of supported parameters in
  LiteLLMCompletionResponsesConfig.get_supported_openai_params()
- Added transformation logic to convert text.format (Responses API format)
  to response_format (Chat Completion API format) in
  transform_responses_api_request_to_chat_completion_request()
- Created _transform_text_format_to_response_format() method to handle
  the conversion between the two format structures

The transformation supports:
- json_schema: Converts to Chat Completion's json_schema format with
  proper nested structure
- json_object: Converts to Chat Completion's json_object mode
- text: Returns None (default text format)

This fix benefits all providers that use the fallback transformation
handler, ensuring consistent behavior with OpenAI's native Responses API
implementation.
2025-10-28 18:06:48 -07:00
Ishaan Jaffer
b0a2e08a60 fixes test 2025-10-28 17:43:04 -07:00
Ishaan Jaffer
d32890ba55 fix _redact_base64 2025-10-28 17:38:16 -07:00
Ishaan Jaffer
1b49dba1dd fix claude-sonnet-4-5 2025-10-28 17:37:08 -07:00
Ishaan Jaffer
74e4d3f6da fixes for mock tests 2025-10-28 17:31:54 -07:00
Ishaan Jaffer
23b1f1afda fix _process_messages 2025-10-28 16:47:52 -07:00
Ishaan Jaffer
cf78b3464c fix linting 2025-10-28 16:46:44 -07:00
Sameer Kankute
8f2becd1c4
Fix: Redact reasoning summaries in ResponsesAPI output when message logging is disabled (#15965)
* redact reasoning content as well

* fix mypy error
2025-10-28 16:42:41 -07:00
Sameer Kankute
25f12924c4
Fix deletion of original request (#16002) 2025-10-28 16:42:08 -07:00
Ishaan Jaff
ab8a3a5d9e
[Fix] SQS Logger - Add Base64 handling (#16028)
* Enable base64 stripping from sqs (#15927)

* Add sqs logger

* Add sqs logger

* Add sqs strp base64

* Add sqs strp base64

* Add sqs strp base64

* strip base64

* Add sqs strp base64

* strip base64

* Add sqs strp base64

* Add max depth recursion

* Add max depth recursion

---------

Co-authored-by: deepanshu <deepanshu.lulla@hq.bill.com>

* refactor _strip_base64_from_messages

* test fixes SQS logger

* fix SQS linting

---------

Co-authored-by: Deepanshu Lulla <deepanshu.lulla@gmail.com>
Co-authored-by: deepanshu <deepanshu.lulla@hq.bill.com>
2025-10-28 16:41:32 -07:00
Ishaan Jaff
95dd216150
[UI] Feature - Add Apply Guardrail Testing Playground (#16030)
* add applyGuardrail endpoints

* v0 testing apply guard

* fix: use tabs

* move apply guardrails endpoint

* fix apply_guardrail

* fix applyGuardrail

* fix apply guardrail for bedrock

* test guard endpoints

* add tooltip for enter button

* refactor

* add guardrail test

* tests guardrails selector

* TestNomaApplyGuardrail
2025-10-28 16:41:17 -07:00
Ishaan Jaff
5c375b23ae
[Fix] Guardrails - Ensure Key Guardrails are applied (#16025)
* _add_guardrails_from_key_or_team_metadata

* test_team_guardrails_append_to_key_guardrails

* fix move_guardrails_to_metadata

* fix _add_guardrails_from_key_or_team_metadata
2025-10-28 16:40:49 -07:00
yuneng-jiang
12de66dad6
Config Models should not be editable (#16020) 2025-10-28 15:27:10 -07:00
Sameer Kankute
59189c0579
fix errors in videos documentation (#15996) 2025-10-28 14:48:04 -07:00
Deepanshu Lulla
3a7c498eff
Add GitlabPromptCache and enable subfolder access (#15712)
* Add GitlabPromptCache and enable subfolder access

* Add GitlabPromptCache and enable subfolder access

* Add GitlabPromptCache and enable subfolder access

---------

Co-authored-by: deepanshu <deepanshu.lulla@hq.bill.com>
2025-10-28 14:46:50 -07:00
Ariel
647f2f5d86
[feat]: graceful degradation for pillar service when using litellm (#15857)
* graceful degradation for pillar service when using litellm

* remove unnecessary mode

* simplify docs

* final fixes

* lint fixes

* fix linting
2025-10-27 19:51:29 -07:00
Thomas Mildner
e27bab3238
fix(opik): enhance requester metadata retrieval from API key auth (#15897) 2025-10-27 19:48:40 -07:00
Thomas Schmidt
2e7dc56895
Add Haiku 4.5 pricing for open router (#15909)
* Add Haiku 4.5 pricing for open router

* Add haiku 4.5 pricing for open router
2025-10-27 19:47:50 -07:00
Katsuhiro Muto
2074b4d662
Fix: Support tool usage messages with Langfuse OTEL integration (#15932)
* Log tool use in langfuse otel integration

* Add test for logging function calling

---------

Co-authored-by: eycjur <eycjur@example.com>
2025-10-27 19:47:31 -07:00
Chris Gibbons
2bef7c3662
fix: Preserve Bedrock inference profile IDs in health checks (#15947)
* fix: Preserve Bedrock inference profile IDs in health checks

- Fixes issue where health checks were stripping inference profile IDs
- Preserves cross-region inference profile prefixes (us., eu., apac., jp., au., us-gov., global.)
- Strips only AWS region routing while preserving routes and handlers
- Resolves both issue #15807 and inference profile requirement errors
- Adds comprehensive tests for all Bedrock model format combinations

Issue #15807 attempted to fix regional Bedrock model health checks but was too
aggressive, stripping cross-region inference profile prefixes that AWS requires.
This caused errors: "Invocation of model ID X with on-demand throughput isn't
supported. Retry your request with the ID or ARN of an inference profile."

The fix now correctly:
- Strips AWS regions (us-west-2, eu-central-1, etc.) from routing
- Preserves CRIS prefixes (us., eu., etc.) required by AWS
- Preserves routes (converse/, invoke/)
- Preserves handlers (llama/, deepseek_r1/)
- Only affects Bedrock models (checked via startswith)

Test coverage includes 20+ scenarios for all Bedrock model format combinations.

* Remove unused traceback import
2025-10-27 19:44:45 -07:00
YutaSaito
8b33328cc1
Perf speed up pytest (#15951)
* perf: Skip sleep delays in base_mail.py during tests to improve test speed

* perf: Mock datetime.now in parallel_request_limiter_v3.py to improve test speed

* pref: Mock urllib system calls in test_aiohttp_transport.py to improve test speed

* chore: add --durations=50 to visualize slowest tests

* pref: reduce setup phase overhead by widening fixture scope in conftest.py

* test: stabilize flaky tests

* fix: minor issue
2025-10-27 19:43:40 -07:00
Mac Misiura
5ad108bc9b
📝 updated ibm_guardrails.md to better indicate how detectors could be configured (#15971) 2025-10-27 19:41:10 -07:00
dima-hx430
c5c37bf7f5
Add models missing deprecation dates (#15976) 2025-10-27 19:39:38 -07:00
Ishaan Jaff
4cef208c5f
[Fix] - Responses API - add /openai routes for responses API. (Azure OpenAI SDK Compatibility) (#15988)
* add /openai routes for responses API

* TestResponsesAPIEndpoints
2025-10-27 19:12:13 -07:00
yuneng-jiang
43af45ab88
Key Already Exist Error Notification (#15993) 2025-10-27 18:00:38 -07:00
Ishaan Jaffer
a3d64fb843 fix omni-moderation-latest 2025-10-27 17:48:35 -07:00
Ishaan Jaffer
0e23f89eb7 fix ModelArmorGuardrail 2025-10-27 17:47:19 -07:00
Ishaan Jaffer
d5f48c7e23 get_metadata_variable_name_from_kwargs 2025-10-27 17:38:24 -07:00
Ishaan Jaffer
de6fffe743 bump: version 1.79.0 → 1.79.1 2025-10-27 17:22:54 -07:00
yuneng-jiang
64167b7e34
Remove limit from admin UI numerical input fix (#15991) 2025-10-27 17:20:58 -07:00
yuneng-jiang
4ed9c7d7f2
[Feat] UI - Changed API Base from Select to Input in New LLM Credentials (#15987)
* Changed API Base from Select to Input

* Added Tests
2025-10-27 15:46:04 -07:00
Ishaan Jaffer
1acc321eb3 test_router_amoderation 2025-10-27 13:50:32 -07:00
Ishaan Jaffer
2d836dfb6d test_basic_moderations_on_proxy_with_model 2025-10-27 13:49:47 -07:00
Ishaan Jaffer
cb57455172 test_foward_litellm_user_info_to_backend_llm_call 2025-10-27 13:48:23 -07:00
Sameer Kankute
59df75276c
Fix: Respect LiteLLM-Disable-Message-Redaction header for Responses API (#15966)
* fix overide for logging unredacted messages

* Use _get_metadata_variable_name_from_kwargs

* fix test related to redaction
2025-10-27 13:46:21 -07:00
Ishaan Jaff
0bb53f5048
[Fix] Azure OpenAI - Add handling for v1 under azure api versions (#15984)
* fix _is_azure_v1_api_version

* test_is_azure_v1_api_version
2025-10-27 13:45:44 -07:00
Ishaan Jaff
65afdda3ef
fix exception triggered only when not logging (#15982) 2025-10-27 13:45:34 -07:00
Ishaan Jaff
17f6238d2b
[Feat] OTEL - Ensure error information is logged on OTEL (#15978)
* fix _record_exception_on_span

* _record_exception_on_span

* test_record_exception_on_span

* fix linting errors
2025-10-27 13:45:21 -07:00
Ishaan Jaff
02df4c6b30
[Fix] DD logging - ensure key's metadata + guardrail is logged on DD (#15980)
* fix get_sanitized_user_information_from_key

* test_get_sanitized_user_information_from_key_includes_guardrails_metadata
2025-10-27 13:45:09 -07:00
Ishaan Jaff
20f9e189fb
[Buf fix] - Azure OpenAI, fix ContextWindowExceededError is not mapped from Azure openai errors (#15981)
* fix is_error_str_context_window_exceeded

* test_is_error_str_context_window_exceeded
2025-10-27 13:44:55 -07:00
Shanto Mathew
4535b5847d
docs(openrouter): add base_url config with environment variables (#15946)
Added a new "Configuration with Environment Variables" section demonstrating:
- Using os.getenv() to dynamically retrieve OpenRouter configuration
- Explicitly passing base_url parameter with environment variables
- Benefits of this approach for managing configs across environments

This helps users implement production-ready configuration patterns.
2025-10-27 13:35:52 -07:00
pinkgu
4758e2998b
feat: support during_call for model armor guardrails (#15970)
Signed-off-by: bjornjee <bjornjee95@gmail.com>
2025-10-27 13:34:59 -07:00
Rob Geada
49b9bd3cad
Update IBM Guardrails implementation to correctly registrer SSL Verify argument (#15975)
Signed-off-by: Rob Geada <rob@geada.net>
2025-10-27 13:00:32 -07:00
Krrish Dholakia
e1f54ef02c docs: refactor placement of adding guardrails to endpoints doc 2025-10-27 10:26:00 -07:00
YutaSaito
c0890e7d33
[Feat] add support for dynamic client registration (#15921) (enables Atlassian MCP to work via Oauth on LiteLLM)
* feat: add support for dynamic client registration #13856

* fix: test

* feat: return 401 when oauth2_header is missing for OAuth2-based MCP servers
2025-10-26 10:13:46 -07:00