Update app.py

app.py

@@ -525,8 +525,9 @@ def get_response_from_llama(query, model, selected_docs, file_type, num_calls=1,
             # Generate content with streaming enabled
             for message in client.chat_completion(
                 messages=messages,
-                max_tokens=20000,
+                max_tokens=3000,
                 temperature=temperature,
+                stream=True,
     
             ):
                 if message.choices and message.choices[0].delta and message.choices[0].delta.content: