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

fix(utils.py): return last streaming chunk

Browse source
This commit is contained in:
Krrish Dholakia 2023-11-29 12:11:08 -08:00
parent a9ed768991
commit 2b437a2699
2 changed files with 8 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

10
litellm/utils.py
View file

@ -5242,6 +5242,11 @@ class CustomStreamWrapper:
return model_response
else:
return
elif model_response.choices[0].finish_reason:
model_response.choices[0].finish_reason = map_finish_reason(model_response.choices[0].finish_reason) # ensure consistent output to openai
# LOGGING
threading.Thread(target=self.logging_obj.success_handler, args=(model_response,)).start()
return model_response
elif response_obj is not None and response_obj.get("original_chunk", None) is not None: # function / tool calling branch - only set for openai/azure compatible endpoints
# enter this branch when no content has been passed in response
original_chunk = response_obj.get("original_chunk", None)
@ -5263,11 +5268,6 @@ class CustomStreamWrapper:
self.sent_first_chunk = True
threading.Thread(target=self.logging_obj.success_handler, args=(model_response,)).start() # log response
return model_response
elif model_response.choices[0].finish_reason:
model_response.choices[0].finish_reason = map_finish_reason(model_response.choices[0].finish_reason) # ensure consistent output to openai
# LOGGING
threading.Thread(target=self.logging_obj.success_handler, args=(model_response,)).start()
return model_response
else:
return
except StopIteration: