fix(main.py): return async completion calls

litellm/llms/openai.py

@@ -631,6 +631,7 @@ class OpenAITextCompletion(BaseLLM):
                         api_key: str, 
                         model: str): 
         async with httpx.AsyncClient() as client:
+            try: 
                 response = await client.post(api_base, json=data, headers=headers, timeout=litellm.request_timeout) 
                 response_json = response.json()
                 if response.status_code != 200:
@@ -649,6 +650,8 @@ class OpenAITextCompletion(BaseLLM):
 
                 ## RESPONSE OBJECT
                 return self.convert_to_model_response_object(response_object=response_json, model_response_object=model_response)
+            except Exception as e: 
+                raise e
 
     def streaming(self,
                   logging_obj,
@@ -687,9 +690,12 @@ class OpenAITextCompletion(BaseLLM):
                     method="POST",
                     timeout=litellm.request_timeout
                 ) as response: 
+            try: 
                 if response.status_code != 200:
                     raise OpenAIError(status_code=response.status_code, message=response.text)
                 
                 streamwrapper = CustomStreamWrapper(completion_stream=response.aiter_lines(), model=model, custom_llm_provider="text-completion-openai",logging_obj=logging_obj)
                 async for transformed_chunk in streamwrapper:
                     yield transformed_chunk
+            except Exception as e: 
+                raise e

litellm/main.py

@@ -2205,7 +2205,8 @@ def text_completion(
     if stream == True or kwargs.get("stream", False) == True:
         response = TextCompletionStreamWrapper(completion_stream=response, model=model)
         return response
-
+    if kwargs.get("acompletion", False) == True: 
+        return response
     transformed_logprobs = None
     # only supported for TGI models
     try:

litellm/tests/test_text_completion.py

@@ -169,17 +169,37 @@ def test_text_completion_stream():
 
 # test_text_completion_stream()
 
-async def test_text_completion_async_stream():
-    try:
-        response = await atext_completion(
-                model="text-completion-openai/text-davinci-003", 
-                prompt="good morning",
-                stream=True,
-                max_tokens=10,
-            )
-        async for chunk in response:
-            print(f"chunk: {chunk}")
-    except Exception as e:
-        pytest.fail(f"GOT exception for HF In streaming{e}")
+# async def test_text_completion_async_stream():
+#     try:
+#         response = await atext_completion(
+#                 model="text-completion-openai/text-davinci-003", 
+#                 prompt="good morning",
+#                 stream=True,
+#                 max_tokens=10,
+#             )
+#         async for chunk in response:
+#             print(f"chunk: {chunk}")
+#     except Exception as e:
+#         pytest.fail(f"GOT exception for HF In streaming{e}")
 
-asyncio.run(test_text_completion_async_stream())
+# asyncio.run(test_text_completion_async_stream())
+        
+def test_async_text_completion():
+    litellm.set_verbose = True
+    print('test_async_text_completion')
+    async def test_get_response():
+        try:
+            response = await litellm.atext_completion(
+                model="gpt-3.5-turbo-instruct", 
+                prompt="good morning",
+                stream=False,
+                max_tokens=10
+            )
+            print(f"response: {response}")
+        except litellm.Timeout as e: 
+            print(e)
+        except Exception as e: 
+            print(e)
+
+    asyncio.run(test_get_response())
+test_async_text_completion()