scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-09 22:31:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 87
25841 commits 4181 branches 1403 tags 2.3 GiB
Commit graph

25841 commits

This branch
This branch
All branches
Author SHA1 Message Date
Ishaan Jaff
69a464fc97
[Fix Security] Ensure OCI secret fields not shared on /models and /v1/models endpoints (#15085) 
* fix: remove_sensitive_info_from_deployment

* fix: remove_sensitive_info_from_deployment

* test_model_info_v1_oci_secrets_not_leaked
 2025-09-30 14:55:54 -07:00
Ishaan Jaffer
75d22d3d79 fix code qa check 2025-09-30 14:03:05 -07:00
Copilot
71b9b58fa9
[Feature]: Replace HTTPException with ParallelRequestLimitError in parallel_request_limiter_v3 (#15033) 
* Initial plan

* Implement ParallelRequestLimitError custom exception to replace HTTPException

Co-authored-by: ishaan-jaff <29436595+ishaan-jaff@users.noreply.github.com>

* Add ParallelRequestLimitError to litellm main module exports

Co-authored-by: ishaan-jaff <29436595+ishaan-jaff@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: ishaan-jaff <29436595+ishaan-jaff@users.noreply.github.com>
 2025-09-30 13:57:14 -07:00
Alexsander Hamir
05dd104ce6
perf(router): Cache nested dict lookups in hot path (#15084) 
Cache deployment["litellm_params"] and deployment["model_info"]
at loop start to avoid repeated dict hash lookups.

- _pre_call_checks: 3 fewer lookups per deployment per request
- deployment_callback_on_failure: 1 fewer lookup per failure
- _set_model_group_info: 4 fewer lookups per model

Saves CPU cycles on every routing decision and failure callback.
 2025-09-30 13:36:42 -07:00
Alexsander Hamir
927e15996e
perf(router): Remove unnecessary hasattr checks in get_model_list() (#15082) 
Remove redundant hasattr() checks for model_list and model_group_alias
in get_model_list() method. Both attributes are always initialized in
__init__, making these runtime checks unnecessary overhead.

Changes:
- Remove hasattr(self, "model_list") check
- Remove hasattr(self, "model_group_alias") check
- Move model_group_alias initialization earlier in __init__ to ensure
  it's available when set_model_list() calls get_model_names()
- Simplify control flow by removing nested conditional blocks

Performance Impact:
- hasattr should not appear in the profile of a proxy server handling thousands of requests. This change ensures it no longer does.
 2025-09-30 13:35:27 -07:00
Ishaan Jaff
60230e5666
[Feat] UI - add snowflake on UI (#15083) 
* UI - add snowflake on UI

* fixes snowflake creds
 2025-09-30 13:16:04 -07:00
Ishaan Jaff
862736e74b
feat: add groq/moonshotai/kimi-k2-instruct-0905 (#15079) 2025-09-30 12:51:21 -07:00
Krish Dholakia
39d1d2a379
Merge pull request #14840 from eddierichter-amd/lemonade-integration 
Add AMD Lemonade provider support
 2025-09-30 12:34:30 -07:00
Ishaan Jaffer
45bfd2599c docs fix 2025-09-30 11:35:15 -07:00
Eddie Richter
7da05df534 Removing unecessary import 2025-09-30 12:12:24 -06:00
Eddie Richter
19e7070b73 Removing get_model_info from Lemonade provider. Implemented get_models which gets hooked into get_valid_models litellm utility. Also, added a simple cost calculator implementation for Lemonade so calling cost_calculator.completion_cost() doesn't return an error when a model is not found in the model_cost json. 2025-09-30 12:12:24 -06:00
Eddie Richter
1e1e4c36ac Fixing key name 2025-09-30 12:12:24 -06:00
Eddie Richter
bbfa00c61b fixing mypy lint errors 2025-09-30 12:12:24 -06:00
Eddie Richter
3d62596daa fix lint-ruff 2025-09-30 12:12:24 -06:00
Eddie Richter
abaf77da43 Small updates to documentation 2025-09-30 12:12:24 -06:00
Eddie Richter
929510ef5d Adding unit tests and documentation 2025-09-30 12:12:24 -06:00
Eddie Richter
6916f43843 Adding functionality for Lemonade to check to see if it is aware of a model and if so use that model 2025-09-30 12:12:24 -06:00
Eddie Richter
351b63bc67 Adding max_tokens to constructor of LemonadeChatConfig 2025-09-30 12:12:24 -06:00
Eddie Richter
f9e98f75a6 Adding max_input_tokens and max_output_tokens 2025-09-30 12:12:24 -06:00
Eddie Richter
0e045e0bb7 Setting the response model so the cost can be calculated 2025-09-30 12:12:24 -06:00
Eddie Richter
eb71611a97 Adding lemonade transform 2025-09-30 12:12:24 -06:00
Eddie Richter
ae92404d05 Initial addition of Lemonade provider. 2025-09-30 12:12:24 -06:00
Krish Dholakia
fd1a986596
Merge pull request #15013 from serializer/patch-1 
[bug]: Update request handling for original exceptions
 2025-09-29 21:59:26 -07:00
Krish Dholakia
79d071c25c
Merge pull request #15022 from BerriAI/litellm_gemini_tool_search 
Ignore type param for gemini tools
 2025-09-29 21:57:50 -07:00
Krish Dholakia
db62e33a4f
Merge pull request #15025 from Isydmr/main 
fix: use extra_query for download results (Batch API)
 2025-09-29 21:57:15 -07:00
Krish Dholakia
551a171243
Merge pull request #15043 from cedarm/fix-remove-vertex-latest 
fix: remove invalid vertex -latest models
 2025-09-29 21:54:53 -07:00
Krish Dholakia
9a44fcdc32
Merge pull request #15050 from uc4w6c/fix/regression-mcp-protocol-version 
fix: resolve regression with duplicate Mcp-Protocol-Version header
 2025-09-29 21:53:00 -07:00
Krish Dholakia
56fccee731
Merge pull request #15058 from uc4w6c/doc/add_api_key_to_bedrock 
doc: add missing api_key parameter
 2025-09-29 21:52:31 -07:00
Yuta Saito
def4afedd7 doc: add missing api_key parameter 2025-09-30 12:52:17 +09:00
Ishaan Jaff
708c0bd78d
[Feat] Return Cost for Responses API Streaming requests (#15053) 
* test_basic_openai_responses_api_streaming

* _transform_chat_completion_usage_to_responses_usage

* ResponseAPIUsage.cost

* test fixes for anthropic cost with /responses

* fix mypy typng
 2025-09-29 19:47:04 -07:00
Ishaan Jaffer
3e474b9e81 fix claude-sonnet-4-5 model cost map 2025-09-29 18:26:29 -07:00
Ishaan Jaffer
f1578b49e2 vertex_httpx_mock_post 2025-09-29 18:25:54 -07:00
Copilot
f22fd4cddd
Fix: Add /v1/messages/count_tokens to Anthropic routes for non-admin user access (#15034) 
* Initial plan

* Fix: Add /v1/messages/count_tokens to Anthropic routes for user access

Co-authored-by: ishaan-jaff <29436595+ishaan-jaff@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: ishaan-jaff <29436595+ishaan-jaff@users.noreply.github.com>
 2025-09-29 18:16:52 -07:00
Ishaan Jaff
ebf72f5eb9
[Fix] Parallel Request Limiter v3 - use well known redis cluster hashing algorithm (#15052) 
* test_keyslot_for_redis_cluster

* fix _is_redis_cluster

* Update litellm/proxy/hooks/parallel_request_limiter_v3.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
 2025-09-29 18:12:44 -07:00
Ishaan Jaff
f6d7683261
[Feat] LiteLLM Overhead metric tracking - Add support for tracking litellm overhead on cache hits (#15045) 
* test_litellm_overhead

* vertex track overhead

* fix config.yaml used for testing

* test_litellm_overhead_stream

* add update_response_metadata for caching handler

* add CachingDetails

* fix update_response_metadata import

* add CachingDetails metrics

* add CachingDetails

* test_litellm_overhead_cache_hit

* test_litellm_overhead_cache_hit

* test_litellm_overhead_cache_hit
 2025-09-29 17:33:27 -07:00
Ishaan Jaffer
55110ba6ae Revert "fix _is_redis_cluster" 
This reverts commit 67abd8880a.
 2025-09-29 16:42:31 -07:00
Ishaan Jaffer
f303860881 Revert "test_keyslot_for_redis_cluster" 
This reverts commit 52de33787b.
 2025-09-29 16:42:15 -07:00
Ishaan Jaffer
52de33787b test_keyslot_for_redis_cluster 2025-09-29 16:41:32 -07:00
Ishaan Jaffer
67abd8880a fix _is_redis_cluster 2025-09-29 16:40:46 -07:00
Alexsander Hamir
d4830e34e5
fix: remove router inefficiencies (from O(M*N) to O(1)) - 62.5% faster P99 latency (#15046) 
* fix: remove redundant deep copy

set_model_list already does the deep copy at the beginning of the call.

* fix: remove unused model_list arguments

The `model_list` parameter was being passed to classes that did not use it.

* fix: reduce per-request memory and time from O(N×M) to O(N)

No need to create a whole array for a simple look up.

* add: missing test

* fix: remove unused parameter
 2025-09-29 15:49:46 -07:00
Ishaan Jaffer
e0172b86e2 test_litellm_overhead_non_streaming 2025-09-29 15:48:32 -07:00
Yuta Saito
5359a0d6a6 fix: test 2025-09-30 07:25:32 +09:00
Ishaan Jaff
619577d4e8
[Feat] Add litellm overhead metric for VertexAI (#15040) 
* test_litellm_overhead

* vertex track overhead

* fix config.yaml used for testing

* test_litellm_overhead_stream

* add update_response_metadata for caching handler

* Revert "add update_response_metadata for caching handler"

This reverts commit f2a891f2b4.
 2025-09-29 15:15:25 -07:00
Yuta Saito
f1f58bd1d1 fix: resolve regression with duplicate Mcp-Protocol-Version header 2025-09-30 07:12:56 +09:00
Ishaan Jaff
05955042d5
Add model pricing and context window for claude-sonnet-4-5 (#15049) 
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: ishaan <ishaan@berri.ai>
 2025-09-29 15:09:40 -07:00
Sameer Kankute
3ab1c31e4e
(Feat) Add cost tracking for Vertex AI Passthrough /predict endpoint (#15019) 
* Add cost tracking for passthrough for predict endpoint

* restore file
 2025-09-29 13:19:04 -07:00
Ishaan Jaff
038863a1fe
[Feat] Add new claude-sonnet-4-5 model family (#15041) 
* add new claude-sonnet-4-5

* docs fix

* fix tool_use_system_prompt_tokens

* add anthropic.claude-sonnet-4-5-20250929 to bedrock converse models
 2025-09-29 13:09:00 -07:00
Cedar Myers
5357b1e102 fix: remove invalid vertex -latest models 2025-09-29 16:04:33 -04:00
Ishaan Jaff
4dc23e8079
Revert "feat: improve vertex AI/gemini api_base handling for proxy services (…" (#15042) 
This reverts commit ff2d19e4ca.
 2025-09-29 13:03:46 -07:00
Krrish Dholakia
a9dcb51d03 fix: add lint 2025-09-29 12:28:43 -07:00