Update main.py

main.py

@@ -73,7 +73,9 @@ async def generate_response_chunks(prompt: str):
                     await asyncio.sleep(0.01)  # Control streaming speed
             logger.info("Streaming completed.")
 
-        return stream_tokens()
+        # Yield tokens from stream_tokens
+        async for token in stream_tokens():
+            yield token
 
     except Exception as e:
         logger.error(f"Error during generation: {e}")