diff --git a/litellm/main.py b/litellm/main.py
index 820033e50..8dbdb820c 100644
--- a/litellm/main.py
+++ b/litellm/main.py
@@ -165,6 +165,10 @@ async def acompletion(*args, **kwargs):
         if (custom_llm_provider == "openai" 
             or custom_llm_provider == "azure" 
             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 == "huggingface"): # currently implemented aiohttp calls for just azure and openai, soon all. 
             if kwargs.get("stream", False): 
diff --git a/litellm/proxy/proxy_server.py b/litellm/proxy/proxy_server.py
index 470868031..8543a535e 100644
--- a/litellm/proxy/proxy_server.py
+++ b/litellm/proxy/proxy_server.py
@@ -582,6 +582,11 @@ async def chat_completion(request: Request, model: Optional[str] = None):
             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)])
 async def generate_key_fn(request: Request): 
     data = await request.json()
diff --git a/litellm/tests/test_async_fn.py b/litellm/tests/test_async_fn.py
index 920b7524e..69bd97019 100644
--- a/litellm/tests/test_async_fn.py
+++ b/litellm/tests/test_async_fn.py
@@ -56,7 +56,7 @@ def test_async_response_openai():
 
     asyncio.run(test_get_response())
 
-test_async_response_openai()
+# test_async_response_openai()
 
 def test_async_response_azure():
     import asyncio
@@ -130,6 +130,7 @@ def test_get_response_streaming():
 def test_get_response_non_openai_streaming():
     import asyncio
     litellm.set_verbose = True
+    litellm.num_retries = 0
     async def test_async_call():
         user_message = "Hello, how are you?"
         messages = [{"content": user_message, "role": "user"}]
@@ -161,4 +162,4 @@ def test_get_response_non_openai_streaming():
         return response
     asyncio.run(test_async_call())
 
-# test_get_response_non_openai_streaming()
\ No newline at end of file
+test_get_response_non_openai_streaming()
\ No newline at end of file