Fixed await issue

main/api.py

@@ -63,14 +63,11 @@ class InferenceApi(LitAPI):
             stream: bool = False
     ) -> Any:
         """Make an authenticated request to the LLM Server."""
-        base_url = self.llm_config.get('host', 'http://localhost:8002')
-        full_endpoint = f"{base_url.rstrip('/')}/{self._get_endpoint(endpoint).lstrip('/')}"
-
         try:
-            self.logger.info(f"Making {method} request to: {full_endpoint}")
             async with await self._get_client() as client:
                 if stream:
-                    return await client.stream(
+                    # Return the context manager directly, don't await it
+                    return client.stream(
                         method,
                         self._get_endpoint(endpoint),
                         params=params,
@@ -87,7 +84,7 @@ class InferenceApi(LitAPI):
                     return response
 
         except Exception as e:
-            self.logger.error(f"Error in request to {full_endpoint}: {str(e)}")
+            self.logger.error(f"Error in request to {endpoint}: {str(e)}")
             raise
 
     def predict(self, x: str, **kwargs) -> Iterator[str]: