fix(openai.py): switch back to using requests instead of httpx

litellm/__init__.py

@@ -19,7 +19,7 @@ telemetry = True
 max_tokens = 256  # OpenAI Defaults
 drop_params = False
 retry = True
-request_timeout: Optional[float] = 600
+request_timeout: Optional[float] = 6000
 api_key: Optional[str] = None
 openai_key: Optional[str] = None
 azure_key: Optional[str] = None

litellm/llms/openai.py

@@ -4,7 +4,7 @@ import httpx
 from .base import BaseLLM
 from litellm.utils import ModelResponse, Choices, Message, CustomStreamWrapper, convert_to_model_response_object, Usage
 from typing import Callable, Optional
-import aiohttp
+import aiohttp, requests
 import litellm
 
 class OpenAIError(Exception):
@@ -356,7 +356,7 @@ class OpenAIChatCompletion(BaseLLM):
                     additional_args={"complete_input_dict": data},
                 )
             ## COMPLETION CALL
-            response = httpx.post(
+            response = requests.post(
                 api_base, headers=headers, json=data, timeout=litellm.request_timeout
             )
             ## LOGGING
@@ -488,7 +488,6 @@ class OpenAITextCompletion(BaseLLM):
                     url=f"{api_base}",
                     json=data,
                     headers=headers,
-                    timeout=litellm.request_timeout
                 )
                 if response.status_code != 200:
                     raise OpenAIError(status_code=response.status_code, message=response.text)

litellm/tests/test_custom_logger.py

@@ -38,11 +38,6 @@ def test_chat_openai():
         for chunk in response:
             # print(chunk)
             continue
-        response = completion(model="gpt-3.5-turbo",
-                              messages=[{
-                                  "role": "user",
-                                  "content": "Hi 👋 - i'm openai"
-                              }])
         # print(response)
 
     except Exception as e:

litellm/tests/test_model_fallback.py

@@ -1,28 +0,0 @@
-#### What this tests ####
-#    This tests error handling + logging (esp. for sentry breadcrumbs)
-
-import sys, os
-import traceback
-
-sys.path.insert(
-    0, os.path.abspath("../..")
-)  # Adds the parent directory to the system path
-import litellm
-from litellm import embedding, completion
-
-litellm.success_callback = ["posthog"]
-litellm.failure_callback = ["sentry", "posthog"]
-
-litellm.set_verbose = True
-
-model_fallback_list = ["claude-instant-1", "gpt-3.5-turbo", "chatgpt-test"]
-
-user_message = "Hello, how are you?"
-messages = [{"content": user_message, "role": "user"}]
-
-for model in model_fallback_list:
-    try:
-        response = embedding(model="text-embedding-ada-002", input=[user_message])
-        response = completion(model=model, messages=messages)
-    except Exception as e:
-        print(f"error occurred: {traceback.format_exc()}")