phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 02:34:29 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[Bug Fix] caching does not account for thinking or reasoning_effort config (#10140)

Browse source 
* _get_litellm_supported_chat_completion_kwargs

* test caching with thinking
This commit is contained in:
Ishaan Jaff 2025-04-21 22:39:40 -07:00 • committed by GitHub
parent 104e4cb1bc
commit 7cb95bcc96
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 75 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

17
litellm/litellm_core_utils/model_param_helper.py
View file

@ -75,6 +75,10 @@ class ModelParamHelper:
combined_kwargs = combined_kwargs.difference(exclude_kwargs)
return combined_kwargs
@staticmethod
def get_litellm_provider_specific_params_for_chat_params() -> Set[str]:
return set(["thinking"])
@staticmethod
def _get_litellm_supported_chat_completion_kwargs() -> Set[str]:
"""
@ -82,11 +86,18 @@ class ModelParamHelper:
This follows the OpenAI API Spec
"""
all_chat_completion_kwargs = set(
non_streaming_params: Set[str] = set(
getattr(CompletionCreateParamsNonStreaming, "__annotations__", {}).keys()
).union(
set(getattr(CompletionCreateParamsStreaming, "__annotations__", {}).keys())
)
streaming_params: Set[str] = set(
getattr(CompletionCreateParamsStreaming, "__annotations__", {}).keys()
)
litellm_provider_specific_params: Set[str] = (
ModelParamHelper.get_litellm_provider_specific_params_for_chat_params()
)
all_chat_completion_kwargs: Set[str] = non_streaming_params.union(
streaming_params
).union(litellm_provider_specific_params)
return all_chat_completion_kwargs
@staticmethod