Update app.py

app.py

@@ -74,12 +74,12 @@ async def complete(gen:GenModel):
         output = llama.create_chat_completion(
             messages=[
                 {"role": "system", "content": gen.system},
-                {"role": "user", "content": gen.question},
             ],
             temperature=gen.temperature,
             seed=gen.seed,
             #stream=True
         )
+        messages.append({"role": "user", "content": gen.question},)
         print(output)
         """
         for chunk in output:
@@ -94,6 +94,7 @@ async def complete(gen:GenModel):
         """
         et = time()
         output["time"] = et - st
+        messages.append({'role': "assistant", "content": ouput['choices'][0]['message'])
         return output
     except Exception as e:
         logger.error(f"Error in /complete endpoint: {e}")