phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 10:44:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix(azure.py,-openai.py): correctly raise errors if streaming calls fail

Browse source
This commit is contained in:
Krrish Dholakia 2023-12-27 15:08:37 +05:30
parent 9ba520cc8b
commit c9fdbaf898
6 changed files with 110 additions and 24 deletions
Download patch file Download diff file Expand all files Collapse all files

20
litellm/main.py
View file

@ -198,18 +198,16 @@ async def acompletion(*args, **kwargs):
or custom_llm_provider == "ollama"
or custom_llm_provider == "ollama_chat"
or custom_llm_provider == "vertex_ai"
): # currently implemented aiohttp calls for just azure and openai, soon all.
if kwargs.get("stream", False):
response = completion(*args, **kwargs)
): # currently implemented aiohttp calls for just azure, openai, hf, ollama, vertex ai soon all.
init_response = await loop.run_in_executor(None, func_with_context)
if isinstance(init_response, dict) or isinstance(
init_response, ModelResponse
): ## CACHING SCENARIO
response = init_response
elif asyncio.iscoroutine(init_response):
response = await init_response
else:
# Await normally
init_response = await loop.run_in_executor(None, func_with_context)
if isinstance(init_response, dict) or isinstance(
init_response, ModelResponse
): ## CACHING SCENARIO
response = init_response
elif asyncio.iscoroutine(init_response):
response = await init_response
response = init_response
else:
# Call the synchronous function using run_in_executor
response = await loop.run_in_executor(None, func_with_context)