Commit Graph

30881 Commits

Author SHA1 Message Date
Sameer Kankute
22e86cde2a Add support for gemini 3 flash via v1/messages endpoint 2025-12-17 18:03:54 +05:30
Sameer Kankute
ba90985300 Add reasoning effort mapping 2025-12-17 18:03:48 +05:30
Sameer Kankute
8e53d4f3eb Add support for structured output thinkingConfig param 2025-12-17 18:02:15 +05:30
Sameer Kankute
ed6c66c20c Add support for structured output thinkingConfig param 2025-12-17 18:02:15 +05:30
Sameer Kankute
9e3f946a2e Add support for new flash model 2025-12-17 18:02:14 +05:30
Sameer Kankute
969ed1efb8 fix: Vertex AI image edit credential source 2025-12-17 15:56:10 +05:30
Sameer Kankute
085b07cdde Add test for handling gaurdrails param 2025-12-17 14:12:27 +05:30
Sameer Kankute
a1536fdd87 Add conversion of gaurdrail params into bedrock headers 2025-12-17 14:12:27 +05:30
Sameer Kankute
b7550ea2d4 Add test for gaurdrail in passthrough 2025-12-17 12:52:21 +05:30
Sameer Kankute
155532085b Add allm_passthrough_route in _get_pre_call_type 2025-12-17 12:51:46 +05:30
Sameer Kankute
46018bd801 fix: TypeError: TypedDict does not support instance and class checks 2025-12-17 12:47:15 +05:30
Sameer Kankute
2e62dfa319 fix: get_call_types_for_route should return None instead of empty list 2025-12-17 12:46:09 +05:30
Sameer Kankute
4b7589486e Fix: Add handling if call_type is None 2025-12-17 12:43:41 +05:30
Curtis
d1c6eb3a7c
fix(image_edit): add drop_params support and fix Vertex AI config (#18077) 2025-12-17 11:28:34 +05:30
yuneng-jiang
9faee8bba6
Merge pull request #17989 from Jetemple/non-root-ui-path-fix
fix: update UI path handling for non-root Docker and restructure HTML…
2025-12-16 21:57:31 -08:00
Sameer Kankute
d547b43beb
Merge pull request #18107 from BerriAI/litellm_add_gpt_1.5_image
Add gpt-image-1.5-2025-12-16 in model cost map
2025-12-17 11:24:12 +05:30
jk-f5
576580244f
fix(azure_ai): return AzureAnthropicConfig for Claude models in get_provider_chat_config (#18086)
Claude models on Azure AI were incorrectly using AzureAIStudioConfig,
causing tool calls to fail with invalid_request_error because tools
remained in OpenAI format instead of being transformed to Anthropic format.
2025-12-17 11:24:04 +05:30
Xiaohan Fu
aac17e5e22
various fixes for openrouter driven models (/messages and its streaming) and introduce the option to send all chunked responses in stream towards the guardrail. (#18085) 2025-12-17 11:21:26 +05:30
Sameer Kankute
8f7ea945f1 add cached token pricing 2025-12-17 11:20:00 +05:30
Shivam Rawat
14a4a9c031
added extraction of top level metadata for custom lables in prometheus callbacks (#18087) 2025-12-17 11:17:12 +05:30
Sameer Kankute
fabd832a39 Add gpt-image-1.5-2025-12-16 in model cost map 2025-12-17 11:16:22 +05:30
Kris Xia
be2f429087
fix(proxy): extract model from vertex ai passthrough url pattern (#18097)
extract model id from vertex ai passthrough routes that follow the pattern:
/vertex_ai/*/models/{model_id}:*

the model extraction now handles vertex ai routes by regex matching the model
segment from the url path, which allows proper model identification for
authentication and authorization in proxy pass-through endpoints.

adds comprehensive test coverage for vertex ai model extraction including:
- various vertex api versions (v1, v1beta1)
- different locations (us-central1, asia-southeast1)
- model names with special suffixes (gemini-1.5-pro, gemini-2.0-flash)
- precedence verification (request body model over url)
- non-vertex route isolation
2025-12-17 11:10:14 +05:30
Cesar Garcia
089d1eb08b
fix(docker): add libsndfile to Alpine image for audio processing (#18092)
ARM64 Alpine image was missing libsndfile library causing soundfile
module to fail with "cannot load library" error.
2025-12-17 10:07:57 +05:30
Ishaan Jaff
0e2fd78ccc
Update GCS bucket logging links to new observability docs (#18104)
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: ishaan <ishaan@berri.ai>
2025-12-17 08:18:53 +04:00
mubashir1osmani
ad91d65766 add 2025-12-16 23:07:32 -05:00
mubashir1osmani
8c2d9e322c fix: arize nested spans 2025-12-16 23:00:56 -05:00
Sameer Kankute
f83b821f3d fix: Add qwen 2 and qwen 3 in get_bedrock_model_id 2025-12-17 09:22:55 +05:30
yuneng-jiang
58330f852d WIP waiting for okta 2025-12-16 17:36:08 -08:00
yuneng-jiang
42d7d757a3 Adding role mappings to SSOConfig DB 2025-12-16 15:48:26 -08:00
yuneng-jiang
9274860aa2 Base commit 2025-12-16 14:54:26 -08:00
yuneng-jiang
1c5f16fb6e
Merge pull request #18034 from BerriAI/litellm_ui_spend_logs_overhead
[Feature] UI -  Show LiteLLM Overhead in Logs
2025-12-16 14:52:26 -08:00
yuneng-jiang
f316caa088
Merge pull request #18033 from BerriAI/litellm_ui_logs_overhead
[Feature] Add LiteLLM Overhead to Logs
2025-12-16 14:51:50 -08:00
Ishaan Jaffer
cdcbccb30d fix interactions API 2025-12-17 02:28:38 +04:00
Ishaan Jaffer
53b47289ce init interactions API 2025-12-17 02:27:54 +04:00
Ishaan Jaffer
f80628502d init docs 2025-12-17 02:27:54 +04:00
Ishaan Jaffer
0052953fc9 docs init 2025-12-17 02:27:54 +04:00
YutaSaito
20cd637f74
Merge pull request #18082 from BerriAI/docs/vault_mount
docs: add documentation describing configurable Hashicorp Vault mount…
2025-12-17 07:24:34 +09:00
Jorge Yero Salazar
48a3a741e5
Allow base_model for non Azure providers in proxy (#18038)
* Allow base_model for non Azure providers in proxy

* Add tests
2025-12-17 02:24:12 +04:00
Yuta Saito
c3d6f8b900 docs: add documentation describing configurable Hashicorp Vault mount and path options 2025-12-17 07:23:30 +09:00
Ishaan Jaff
424363fb1d
[Feat] Add New Google Interactions API on AI Gateway (#18081)
* init BaseInteractionsAPIConfig

* updated types for interactions API

* fix SDK level interactions API

* InteractionsAPIRequestUtils

* init base

* init interactions API

* init Interactions API Types

* init interactins API

* GoogleAIStudioInteractionsConfig

* init http handler

* remove file no longer needed

* test OPENAPI_SPEC_URL

* TestGoogleInteractionsCreate

* add google interctions routes

* init interactions api

* init interactions api

* new routes

* fix config

* working streaming interactons api
2025-12-17 02:20:23 +04:00
Ishaan Jaff
c727c8216f
[Feat] New Endpoint - Google Interactions API - added on litellm SDK (#18079)
* init BaseInteractionsAPIConfig

* updated types for interactions API

* fix SDK level interactions API

* InteractionsAPIRequestUtils

* init base

* init interactions API

* init Interactions API Types

* init interactins API

* GoogleAIStudioInteractionsConfig

* init http handler

* remove file no longer needed

* test OPENAPI_SPEC_URL

* TestGoogleInteractionsCreate
2025-12-17 02:18:51 +04:00
codgician
58e081b6e0
feat: add github_copilot model info (#17858) 2025-12-17 02:15:13 +04:00
yuneng-jiang
c2f79681b6 Fixing test 2 2025-12-16 13:44:53 -08:00
yuneng-jiang
2d8f5111af add /sso/readiness route 2025-12-16 13:16:13 -08:00
yuneng-jiang
dc79da871e Base commit 2025-12-16 13:15:32 -08:00
Emerson Gomes
8df713b887
Add Azure DeepSeek V3.2 versions (#18019) 2025-12-17 01:01:57 +04:00
yuneng-jiang
0e5b56eddf Fixing tests 2025-12-16 12:51:44 -08:00
Ishaan Jaff
77b8004ab1
add brazil pii patterns (#18076) 2025-12-17 00:47:10 +04:00
Alexsander Hamir
233e89976a
Lazy load OpenAILikeChatConfig to avoid heavy import (#18075)
- Remove direct import of OpenAILikeChatConfig from llms.openai_like.chat.handler
- Add OpenAILikeChatConfig to LLM_CONFIG_NAMES for lazy loading
- Add handler in _lazy_import_llm_configs() to lazy load when accessed
- Add OpenAILikeChatConfig to TYPE_CHECKING block for type hints

This defers the import until OpenAILikeChatConfig is actually needed, improving import time.
2025-12-16 12:43:42 -08:00
Alexsander Hamir
9b440caac2
Lazy load ImageEditRequestUtils in images/main.py to avoid heavy import (#18074)
- Remove direct import of ImageEditRequestUtils from images.utils
- Add __getattr__ handler to lazy load ImageEditRequestUtils when accessed
- Create _get_ImageEditRequestUtils() helper function with caching
- Use importlib instead of sys to avoid importing sys inside functions
- Update image_edit() to use helper function for accessing ImageEditRequestUtils
- Add TYPE_CHECKING import for proper type hints

This defers the heavy import from images.utils until ImageEditRequestUtils is actually needed, improving import time when from .images.main import * is executed.
2025-12-16 12:28:33 -08:00