phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 18:54:30 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

handle _get_async_http_client for OpenAI

Browse source
This commit is contained in:
Ishaan Jaff 2025-03-18 08:56:08 -07:00
parent d42f84dd4e
commit 6e351136d7
1 changed files with 24 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

26
litellm/llms/openai/openai.py
View file

@ -370,7 +370,7 @@ class OpenAIChatCompletion(BaseLLM):
_new_client: Union[OpenAI, AsyncOpenAI] = AsyncOpenAI(
api_key=api_key,
base_url=api_base,
http_client=litellm.aclient_session,
http_client=OpenAIChatCompletion._get_async_http_client(),
timeout=timeout,
max_retries=max_retries,
organization=organization,
@ -379,7 +379,7 @@ class OpenAIChatCompletion(BaseLLM):
_new_client = OpenAI(
api_key=api_key,
base_url=api_base,
http_client=litellm.client_session,
http_client=OpenAIChatCompletion._get_sync_http_client(),
timeout=timeout,
max_retries=max_retries,
organization=organization,
@ -401,6 +401,28 @@ class OpenAIChatCompletion(BaseLLM):
)
return client
@staticmethod
def _get_async_http_client() -> Optional[httpx.AsyncClient]:
if litellm.ssl_verify:
return httpx.AsyncClient(
limits=httpx.Limits(
max_connections=1000, max_keepalive_connections=100
),
verify=litellm.ssl_verify,
)
return litellm.aclient_session
@staticmethod
def _get_sync_http_client() -> Optional[httpx.Client]:
if litellm.ssl_verify:
return httpx.Client(
limits=httpx.Limits(
max_connections=1000, max_keepalive_connections=100
),
verify=litellm.ssl_verify,
)
return litellm.client_session
@track_llm_api_timing()
async def make_openai_chat_completion_request(
self,