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

add response ms to response object

Browse source
This commit is contained in:
Krrish Dholakia 2023-09-20 19:50:38 -07:00
parent fc922bc300
commit 39c3f2d6ec
4 changed files with 9 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

7
litellm/utils.py
View file

@ -123,7 +123,7 @@ class StreamingChoices(OpenAIObject):
self.delta = Delta()
class ModelResponse(OpenAIObject):
def __init__(self, id=None, choices=None, created=None, model=None, usage=None, stream=False, **params):
def __init__(self, id=None, choices=None, created=None, model=None, usage=None, stream=False, response_ms=None, **params):
if stream:
self.object = "chat.completion.chunk"
self.choices = [StreamingChoices()]
@ -141,6 +141,10 @@ class ModelResponse(OpenAIObject):
self.created = int(time.time())
else:
self.created = created
if response_ms:
response_ms = response_ms
else:
response_ms = None
self.model = model
self.usage = (
usage
@ -624,6 +628,7 @@ def client(original_function):
) # don't interrupt execution of main thread
my_thread.start()
# RETURN RESULT
result.response_ms = (end_time - start_time).total_seconds() * 1000 # return response latency in ms like openai
return result
except Exception as e:
traceback_exception = traceback.format_exc()