phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 11:14:04 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix - reuse client initialized on proxy config

Browse source
This commit is contained in:
Ishaan Jaff 2024-06-26 16:16:58 -07:00
parent 57852bada9
commit b7bca0af6c
2 changed files with 16 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

3
litellm/llms/azure.py
View file

@ -812,7 +812,7 @@ class AzureChatCompletion(BaseLLM):
azure_client_params: dict, azure_client_params: dict,
api_key: str, api_key: str,
input: list, input: list,
client=None, client: Optional[AsyncAzureOpenAI] = None,
logging_obj=None, logging_obj=None,
timeout=None, timeout=None,
): ):
@ -911,6 +911,7 @@ class AzureChatCompletion(BaseLLM):
model_response=model_response, model_response=model_response,
azure_client_params=azure_client_params, azure_client_params=azure_client_params,
timeout=timeout, timeout=timeout,
client=client,
) )
return response return response
if client is None: if client is None:

18
litellm/llms/openai.py
View file

@ -996,11 +996,11 @@ class OpenAIChatCompletion(BaseLLM):
self, self,
input: list, input: list,
data: dict, data: dict,
model_response: ModelResponse, model_response: litellm.utils.EmbeddingResponse,
timeout: float, timeout: float,
api_key: Optional[str] = None, api_key: Optional[str] = None,
api_base: Optional[str] = None, api_base: Optional[str] = None,
client=None, client: Optional[AsyncOpenAI] = None,
max_retries=None, max_retries=None,
logging_obj=None, logging_obj=None,
): ):
@ -1039,9 +1039,9 @@ class OpenAIChatCompletion(BaseLLM):
input: list, input: list,
timeout: float, timeout: float,
logging_obj, logging_obj,
model_response: litellm.utils.EmbeddingResponse,
api_key: Optional[str] = None, api_key: Optional[str] = None,
api_base: Optional[str] = None, api_base: Optional[str] = None,
model_response: Optional[litellm.utils.EmbeddingResponse] = None,
optional_params=None, optional_params=None,
client=None, client=None,
aembedding=None, aembedding=None,
@ -1062,7 +1062,17 @@ class OpenAIChatCompletion(BaseLLM):
) )
if aembedding is True: if aembedding is True:
response = self.aembedding(data=data, input=input, logging_obj=logging_obj, model_response=model_response, api_base=api_base, api_key=api_key, timeout=timeout, client=client, max_retries=max_retries) # type: ignore response = self.aembedding(
data=data,
input=input,
logging_obj=logging_obj,
model_response=model_response,
api_base=api_base,
api_key=api_key,
timeout=timeout,
client=client,
max_retries=max_retries,
)
return response return response
openai_client = self._get_openai_client( openai_client = self._get_openai_client(