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

fix(utils.py): support raw response headers for streaming requests

Browse source
This commit is contained in:
Krrish Dholakia 2024-07-23 11:58:58 -07:00
parent d1ffb4de5f
commit f64a3309d1
5 changed files with 60 additions and 30 deletions
Download patch file Download diff file Expand all files Collapse all files

43
litellm/tests/test_streaming.py
View file

@ -1988,25 +1988,30 @@ async def test_hf_completion_tgi_stream():
# test on openai completion call
def test_openai_chat_completion_call():
try:
litellm.set_verbose = False
print(f"making openai chat completion call")
response = completion(model="gpt-3.5-turbo", messages=messages, stream=True)
complete_response = ""
start_time = time.time()
for idx, chunk in enumerate(response):
chunk, finished = streaming_format_tests(idx, chunk)
print(f"outside chunk: {chunk}")
if finished:
break
complete_response += chunk
# print(f'complete_chunk: {complete_response}')
if complete_response.strip() == "":
raise Exception("Empty response received")
print(f"complete response: {complete_response}")
except:
print(f"error occurred: {traceback.format_exc()}")
pass
litellm.set_verbose = False
litellm.return_response_headers = True
print(f"making openai chat completion call")
response = completion(model="gpt-3.5-turbo", messages=messages, stream=True)
assert isinstance(
response._hidden_params["additional_headers"][
"llm_provider-x-ratelimit-remaining-requests"
],
str,
)
print(f"response._hidden_params: {response._hidden_params}")
complete_response = ""
start_time = time.time()
for idx, chunk in enumerate(response):
chunk, finished = streaming_format_tests(idx, chunk)
print(f"outside chunk: {chunk}")
if finished:
break
complete_response += chunk
# print(f'complete_chunk: {complete_response}')
if complete_response.strip() == "":
raise Exception("Empty response received")
print(f"complete response: {complete_response}")
# test_openai_chat_completion_call()