Update main.py

main.py

@@ -30,7 +30,7 @@ async def generate_response(item: Validation):
 { item.user_prompt }<|eot_id|> \n <|start_header_id|>assistant<|end_header_id|>"""
     
     # Call the Llama model to generate a response
-    output = llm('Q: what is gravity? A:', max_tokens=1024, stop=["Q:", "\n"], echo=True)  # Update parameters as needed
+    output = llm(prompt, max_tokens=1024, echo=True)  # Update parameters as needed
     
     # Extract and return the text from the response
-    return output
+    return output['choices'][0]['text']