phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-24 18:24:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
21569 commits 2475 branches 894 tags 524 MiB
Commit graph

1660 commits

Author SHA1 Message Date
Ishaan Jaff
c17e5351be test response id 2025-04-21 19:39:34 -07:00
Ishaan Jaff
95337903dd test_responses_api_deployment_check.py 2025-04-21 16:59:14 -07:00
Ishaan Jaff
6967d991f4 test_responses_api_routing_with_previous_response_id 2025-04-21 16:41:08 -07:00
Ishaan Jaff
e82764158e test_responses_api_routing_with_previous_response_id 2025-04-21 14:41:31 -07:00
Ishaan Jaff
6783c9e618 test for test_responses_api_routing_with_previous_response_id 2025-04-21 14:35:47 -07:00
Li Yang
10257426a2
fix(bedrock): wrong system prompt transformation (#10120)
All checks were successful
Read Version from pyproject.toml / read-version (push) Successful in 16s
Details
Helm unit test / unit-test (push) Successful in 25s
Details 
* fix(bedrock): wrong system transformation

* chore: add one more test case

---------

Co-authored-by: Krish Dholakia <krrishdholakia@gmail.com>
 2025-04-21 08:48:14 -07:00
Krish Dholakia
e0a613f88a
fix(common_daily_activity.py): support empty entity id field (#10175) 
* fix(common_daily_activity.py): support empty entity id field

allows returning empty response when user is not admin and does not belong to any team

* test(test_common_daily_activity.py): add unit testing
 2025-04-19 22:20:28 -07:00
Krish Dholakia
55a17730fb
fix(transformation.py): pass back in gemini thinking content to api (#10173) 
Ensures thinking content always returned
 2025-04-19 18:03:05 -07:00
Ishaan Jaff
653570824a
Bug Fix - Responses API, Loosen restrictions on allowed environments for computer use tool (#10168) 
* loosen allowed types on ComputerToolParam

* test_basic_computer_use_preview_tool_call
 2025-04-19 14:40:32 -07:00
Ishaan Jaff
b0024bb229
[Bug Fix] Spend Tracking Bug Fix, don't modify in memory default litellm params (#10167) 
* _update_kwargs_with_default_litellm_params

* test_update_kwargs_does_not_mutate_defaults_and_merges_metadata
 2025-04-19 14:13:59 -07:00
Ishaan Jaff
0717369ae6
[Feat] Expose Responses API on LiteLLM UI Test Key Page (#10166) 
* add /responses API on UI

* add makeOpenAIResponsesRequest

* add makeOpenAIResponsesRequest

* fix add responses API on UI

* fix endpoint selector

* responses API render chunks on litellm chat ui

* fixes to streaming iterator

* fix render responses completed events

* fixes for MockResponsesAPIStreamingIterator

* transform_responses_api_request_to_chat_completion_request

* fix for responses API

* test_basic_openai_responses_api_streaming

* fix base responses api tests
 2025-04-19 13:18:54 -07:00
Krish Dholakia
03b5399f86
test(utils.py): handle scenario where text tokens + reasoning tokens … (#10165) 
* test(utils.py): handle scenario where text tokens + reasoning tokens set, but reasoning tokens not charged separately

Addresses https://github.com/BerriAI/litellm/pull/10141#discussion_r2051555332

* fix(vertex_and_google_ai_studio.py): only set content if non-empty str
 2025-04-19 12:32:38 -07:00
Krish Dholakia
5c929317cd
fix(triton/completion/transformation.py): remove bad_words / stop wor… (#10163) 
* fix(triton/completion/transformation.py): remove bad_words / stop words from triton call

parameter 'bad_words' has invalid type. It should be either 'int', 'bool', or 'string'.

* fix(proxy_track_cost_callback.py): add debug logging for track cost callback error
 2025-04-19 11:23:37 -07:00
Krish Dholakia
f08a4e3c06
Support 'file' message type for VLLM video url's + Anthropic redacted message thinking support (#10129) 
* feat(hosted_vllm/chat/transformation.py): support calling vllm video url with openai 'file' message type

allows switching between gemini/vllm easily

* [WIP] redacted thinking tests (#9044)

* WIP: redacted thinking tests

* test: add test for redacted thinking in assistant message

---------

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

* fix(anthropic/chat/transformation.py): support redacted thinking block on anthropic completion

Fixes https://github.com/BerriAI/litellm/issues/9058

* fix(anthropic/chat/handler.py): transform anthropic redacted messages on streaming

Fixes https://github.com/BerriAI/litellm/issues/9058

* fix(bedrock/): support redacted text on streaming + non-streaming

Fixes https://github.com/BerriAI/litellm/issues/9058

* feat(litellm_proxy/chat/transformation.py): support 'reasoning_effort' param for proxy

allows using reasoning effort with thinking models on proxy

* test: update tests

* fix(utils.py): fix linting error

* fix: fix linting errors

* fix: fix linting errors

* fix: fix linting error

* fix: fix linting errors

* fix(anthropic/chat/transformation.py): fix returning citations in chat completion

---------

Co-authored-by: Johann Miller <22018973+johannkm@users.noreply.github.com>
 2025-04-19 11:16:37 -07:00
Ishaan Jaff
3c463f6715 test fix - output_cost_per_reasoning_token was added to model cost map 2025-04-19 10:02:25 -07:00
Krish Dholakia
2508ca71cb
Handle fireworks ai tool calling response (#10130) 
* feat(fireworks_ai/chat): handle tool calling with fireworks ai correctly

Fixes https://github.com/BerriAI/litellm/issues/7209

* fix(utils.py): handle none type in message

* fix: fix model name in test

* fix(utils.py): fix validate check for openai messages

* fix: fix model returned

* fix(main.py): fix text completion routing

* test: update testing

* test: skip test - cohere having RBAC issues
 2025-04-19 09:37:45 -07:00
Krrish Dholakia
b4f2b3dad1 test: update test to be more robust to usage updates 2025-04-19 09:26:26 -07:00
Ishaan Jaff
8ae2653280 fix calculated cache key for tests 2025-04-19 09:25:11 -07:00
Krish Dholakia
36308a31be
Gemini-2.5-flash - support reasoning cost calc + return reasoning content (#10141) 
* build(model_prices_and_context_window.json): add vertex ai gemini-2.5-flash pricing

* build(model_prices_and_context_window.json): add gemini reasoning token pricing

* fix(vertex_and_google_ai_studio_gemini.py): support counting thinking tokens for gemini

allows accurate cost calc

* fix(utils.py): add reasoning token cost calc to generic cost calc

ensures gemini-2.5-flash cost calculation is accurate

* build(model_prices_and_context_window.json): mark gemini-2.5-flash as 'supports_reasoning'

* feat(gemini/): support 'thinking' + 'reasoning_effort' params + new unit tests

allow controlling thinking effort for gemini-2.5-flash models

* test: update unit testing

* feat(vertex_and_google_ai_studio_gemini.py): return reasoning content if given in gemini response

* test: update model name

* fix: fix ruff check

* test(test_spend_management_endpoints.py): update tests to be less sensitive to new keys / updates to usage object

* fix(vertex_and_google_ai_studio_gemini.py): fix translation
 2025-04-19 09:20:52 -07:00
Krrish Dholakia
d726e0f34c test: update testing imports 2025-04-19 09:13:16 -07:00
Ishaan Jaff
0a35c208d7 test assistants fixes 2025-04-19 08:09:45 -07:00
Ishaan Jaff
a62805f98f fixes for assistans API tests 2025-04-19 07:59:53 -07:00
Ishaan Jaff
5bf76f0bb1 test fixes for azure assistants 2025-04-19 07:36:40 -07:00
Ishaan Jaff
b9756bf006 test_completion_azure 2025-04-19 07:24:11 -07:00
Krrish Dholakia
652e1b7f0f test: update test 2025-04-18 20:36:15 -07:00
Ishaan Jaff
3d5022bd79
[Feat] Support for all litellm providers on Responses API (works with Codex) - Anthropic, Bedrock API, VertexAI, Ollama (#10132) 
* transform request

* basic handler for LiteLLMCompletionTransformationHandler

* complete transform litellm to responses api

* fixes to test

* fix stream=True

* fix streaming iterator

* fixes for transformation

* fixes for anthropic codex support

* fix pass response_api_optional_params

* test anthropic responses api tools

* update responses types

* working codex with litellm

* add session handler

* fixes streaming iterator

* fix handler

* add litellm codex example

* fix code quality

* test fix

* docs litellm codex

* litellm codexdoc

* docs openai codex with litellm

* docs litellm openai codex

* litellm codex

* linting fixes for transforming responses API

* fix import error

* fix responses api test

* add sync iterator support for responses api
 2025-04-18 19:53:59 -07:00
Krrish Dholakia
3e87ec4f16 test: replace removed fireworks ai models
All checks were successful
Read Version from pyproject.toml / read-version (push) Successful in 19s
Details
Helm unit test / unit-test (push) Successful in 24s
Details
2025-04-18 14:23:16 -07:00
Krish Dholakia
1ea046cc61
test: update tests to new deployment model (#10142) 
* test: update tests to new deployment model

* test: update model name

* test: skip cohere rbac issue test

* test: update test - replace gpt-4o model
 2025-04-18 14:22:12 -07:00
Krrish Dholakia
415abfc222 test: update test 2025-04-18 13:13:58 -07:00
Krrish Dholakia
f7dd688035 test: handle cohere rbac issue (verified happens on calling azure directly)
All checks were successful
Read Version from pyproject.toml / read-version (push) Successful in 15s
Details
Helm unit test / unit-test (push) Successful in 23s
Details
2025-04-18 08:42:12 -07:00
Ishaan Jaff
d3e04eac7f
[Feat] Unified Responses API - Add Azure Responses API support (#10116) 
* initial commit for azure responses api support

* update get complete url

* fixes for responses API

* working azure responses API

* working responses API

* test suite for responses API

* azure responses API test suite

* fix test with complete url

* fix test refactor

* test fix metadata checks

* fix code quality check
 2025-04-17 16:47:59 -07:00
Ishaan Jaff
257e78ffb5 test fix vertex_ai/mistral-large@2407 2025-04-16 21:52:52 -07:00
Krish Dholakia
c73a6a8d1e
Add new /vertex_ai/discovery route - enables calling AgentBuilder API routes (#10084) 
* feat(llm_passthrough_endpoints.py): expose new `/vertex_ai/discovery/` endpoint

Allows calling vertex ai discovery endpoints via passthrough

 For agentbuilder api calls

* refactor(llm_passthrough_endpoints.py): use common _base_vertex_proxy_route

Prevents duplicate code

* feat(llm_passthrough_endpoints.py): add vertex endpoint specific passthrough handlers
 2025-04-16 21:45:51 -07:00
Ishaan Jaff
198922b26f test fixes for vertex mistral, this model was deprecated on vertex 2025-04-16 20:51:45 -07:00
Ishaan Jaff
c38146e180 test fix 2025-04-16 20:13:31 -07:00
Ishaan Jaff
cf801f9642 test fix vertex_ai/codestral 2025-04-16 20:01:36 -07:00
Ishaan Jaff
6220f3e7b8
[Feat SSO] Add LiteLLM SCIM Integration for Team and User management (#10072) 
* fix NewUser response type

* add scim router

* add v0 scim v2 endpoints

* working scim transformation

* use 1 file for types

* fix scim firstname and givenName storage

* working SCIMErrorResponse

* working team / group provisioning on SCIM

* add SCIMPatchOp

* move scim folder

* fix import scim_router

* fix dont auto create scim keys

* add auth on all scim endpoints

* add is_virtual_key_allowed_to_call_route

* fix allowed routes

* fix for key management

* fix allowed routes check

* clean up error message

* fix code check

* fix for route checks

* ui SCIM support

* add UI tab for SCIM

* fixes SCIM

* fixes for SCIM settings on ui

* scim settings

* clean up scim view

* add migration for allowed_routes in keys table

* refactor scim transform

* fix SCIM linting error

* fix code quality check

* fix ui linting

* test_scim_transformations.py
 2025-04-16 19:21:47 -07:00
Krish Dholakia
c603680d2a
fix(stream_chunk_builder_utils.py): don't set index on modelresponse (#10063) 
* fix(stream_chunk_builder_utils.py): don't set index on modelresponse

* test: update tests
 2025-04-16 10:11:47 -07:00
Krrish Dholakia
6b7d20c911 test: fix test 2025-04-16 07:57:10 -07:00
Michael Leshchinsky
e19d05980c
Add litellm call id passing to Aim guardrails on pre and post-hooks calls (#10021)
All checks were successful
Read Version from pyproject.toml / read-version (push) Successful in 16s
Details
Helm unit test / unit-test (push) Successful in 19s
Details 
* Add litellm_call_id passing to aim guardrails on pre and post-hooks

* Add test that ensures that pre_call_hook receives litellm call id when common_request_processing called
 2025-04-16 07:41:28 -07:00
Ishaan Jaff
ad09d250ef test fix azure deprecated mistral 2025-04-15 22:32:14 -07:00
Krish Dholakia
fdfa1108a6
Add property ordering for vertex ai schema (#9828) + Fix combining multiple tool calls (#10040) 
* fix #9783: Retain schema field ordering for google gemini and vertex (#9828)

* test: update test

* refactor(groq.py): initial commit migrating groq to base_llm_http_handler

* fix(streaming_chunk_builder_utils.py): fix how tool content is combined

Fixes https://github.com/BerriAI/litellm/issues/10034

* fix(vertex_ai/common_utils.py): prevent infinite loop in helper function

* fix(groq/chat/transformation.py): handle groq streaming errors correctly

* fix(groq/chat/transformation.py): handle max_retries

---------

Co-authored-by: Adrian Lyjak <adrian@chatmeter.com>
 2025-04-15 22:29:25 -07:00
Krish Dholakia
1b9b745cae
Fix gcs pub sub logging with env var GCS_PROJECT_ID (#10042) 
* fix(pub_sub.py): fix passing project id in pub sub call

Fixes issue where GCS_PUBSUB_PROJECT_ID was not being used

* test(test_pub_sub.py): add unit test to prevent future regressions

* test: fix test
 2025-04-15 21:50:48 -07:00
Ishaan Jaff
b3f37b860d test fix azure deprecated mistral ai 2025-04-15 21:42:40 -07:00
Ishaan Jaff
bd88263b29
[Feat - Cost Tracking improvement] Track prompt caching metrics in DailyUserSpendTransactions (#10029) 
* stash changes

* emit cache read/write tokens to daily spend update

* emit cache read/write tokens on daily activity

* update types.ts

* docs prompt caching

* undo ui change

* fix activity metrics

* fix prompt caching metrics

* fix typed dict fields

* fix get_aggregated_daily_spend_update_transactions

* fix aggregating cache tokens

* test_cache_token_fields_aggregation

* daily_transaction

* add cache_creation_input_tokens and cache_read_input_tokens to LiteLLM_DailyUserSpend

* test_daily_spend_update_queue.py
 2025-04-15 21:40:57 -07:00
Ishaan Jaff
d32d6fe03e
[UI] Bug Fix - Show created_at and updated_at for Users Page (#10033) 
* add created_at and updated_at as fields for internal user table

* test_get_users_includes_timestamps
 2025-04-15 21:15:44 -07:00
Ishaan Jaff
65f8015221 test fix - azure deprecated azure ai mistral 2025-04-15 21:08:55 -07:00
Krish Dholakia
d3e7a137ad
Revert "fix #9783: Retain schema field ordering for google gemini and vertex …" (#10038) 
This reverts commit e3729f9855.
 2025-04-15 19:21:33 -07:00
Adrian Lyjak
e3729f9855
fix #9783: Retain schema field ordering for google gemini and vertex (#9828) 2025-04-15 19:12:02 -07:00
Marc Abramowitz
837a6948d8
Fix typo: Entrata -> Entra in code (#9922) 
* Fix typo: Entrata -> Entra

* Fix a few more
 2025-04-15 17:31:18 -07:00