Update main.py

main.py

@@ -23,14 +23,10 @@ app = FastAPI()
 @app.post("/generate_response")
 async def generate_response(item: Validation):
     # Construct the complete prompt using the given system and user prompts
-    prompt = f"""<|begin_of_text|><|start_header_id|>system<|end_header_id|> \n
-
-{ item.system_prompt }<|eot_id|> \n <|start_header_id|>user<|end_header_id|>
-
-{ item.user_prompt }<|eot_id|> \n <|start_header_id|>assistant<|end_header_id|>"""
+    prompt = f"""system \n{item.system_prompt}\nuser \n{item.user_prompt}\n assistant"""
     
     # Call the Llama model to generate a response
-    output = llm(prompt, max_tokens=1024, echo=True)  # Update parameters as needed
+    output = llm(prompt, echo=True)  # Update parameters as needed
     
     # Extract and return the text from the response
     return output['choices'][0]['text']