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

fix(main.py): fix acompletion for anyscale, openrouter, deepinfra, perplexity endpoints

Browse source
This commit is contained in:
Krrish Dholakia 2023-11-22 13:22:43 -08:00
parent ba73224a3a
commit 78582e158a
3 changed files with 12 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
litellm/main.py
View file

@ -165,6 +165,10 @@ async def acompletion(*args, **kwargs):
if (custom_llm_provider == "openai" if (custom_llm_provider == "openai"
or custom_llm_provider == "azure" or custom_llm_provider == "azure"
or custom_llm_provider == "custom_openai" or custom_llm_provider == "custom_openai"
or custom_llm_provider == "anyscale"
or custom_llm_provider == "openrouter"
or custom_llm_provider == "deepinfra"
or custom_llm_provider == "perplexity"
or custom_llm_provider == "text-completion-openai" or custom_llm_provider == "text-completion-openai"
or custom_llm_provider == "huggingface"): # currently implemented aiohttp calls for just azure and openai, soon all. or custom_llm_provider == "huggingface"): # currently implemented aiohttp calls for just azure and openai, soon all.
if kwargs.get("stream", False): if kwargs.get("stream", False):

5
litellm/proxy/proxy_server.py
View file

@ -582,6 +582,11 @@ async def chat_completion(request: Request, model: Optional[str] = None):
detail=error_msg detail=error_msg
) )
@router.post("/v1/embeddings", dependencies=[Depends(user_api_key_auth)])
@router.post("/embeddings", dependencies=[Depends(user_api_key_auth)])
async def embeddings(request: Request):
pass
@router.post("/key/generate", dependencies=[Depends(user_api_key_auth)]) @router.post("/key/generate", dependencies=[Depends(user_api_key_auth)])
async def generate_key_fn(request: Request): async def generate_key_fn(request: Request):
data = await request.json() data = await request.json()

5
litellm/tests/test_async_fn.py
View file

@ -56,7 +56,7 @@ def test_async_response_openai():
asyncio.run(test_get_response()) asyncio.run(test_get_response())
test_async_response_openai() # test_async_response_openai()
def test_async_response_azure(): def test_async_response_azure():
import asyncio import asyncio
@ -130,6 +130,7 @@ def test_get_response_streaming():
def test_get_response_non_openai_streaming(): def test_get_response_non_openai_streaming():
import asyncio import asyncio
litellm.set_verbose = True litellm.set_verbose = True
litellm.num_retries = 0
async def test_async_call(): async def test_async_call():
user_message = "Hello, how are you?" user_message = "Hello, how are you?"
messages = [{"content": user_message, "role": "user"}] messages = [{"content": user_message, "role": "user"}]
@ -161,4 +162,4 @@ def test_get_response_non_openai_streaming():
return response return response
asyncio.run(test_async_call()) asyncio.run(test_async_call())
# test_get_response_non_openai_streaming() test_get_response_non_openai_streaming()