phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

fix(main.py): safely fail stream_chunk_builder calls

Browse source
This commit is contained in:
Krrish Dholakia 2024-08-10 10:22:26 -07:00
parent dd2ea72cb4
commit 068ee12c30
3 changed files with 259 additions and 231 deletions
Download patch file Download diff file Expand all files Collapse all files

9
litellm/utils.py
View file

@ -10307,7 +10307,8 @@ class CustomStreamWrapper:
chunks=self.chunks, messages=self.messages
)
response = self.model_response_creator()
response.usage = complete_streaming_response.usage # type: ignore
if complete_streaming_response is not None:
response.usage = complete_streaming_response.usage
response._hidden_params["usage"] = complete_streaming_response.usage # type: ignore
## LOGGING
threading.Thread(
@ -10504,7 +10505,8 @@ class CustomStreamWrapper:
chunks=self.chunks, messages=self.messages
)
response = self.model_response_creator()
response.usage = complete_streaming_response.usage
if complete_streaming_response is not None:
setattr(response, "usage", complete_streaming_response.usage)
## LOGGING
threading.Thread(
target=self.logging_obj.success_handler,
@ -10544,7 +10546,8 @@ class CustomStreamWrapper:
chunks=self.chunks, messages=self.messages
)
response = self.model_response_creator()
response.usage = complete_streaming_response.usage
if complete_streaming_response is not None:
response.usage = complete_streaming_response.usage
## LOGGING
threading.Thread(
target=self.logging_obj.success_handler,