Has autogen Intrinsic limits?

Hi all

Is there an intrinsic limit imposed by the autogen library to limit the usage of an API key?.

I ran the chess game example in my local machine (using my own API key), but the game stops exactly at the same point it stops in the lecture and in the jupyter notebook. Also if I run the script again, the response is almost instantaneous (as it came from a cache) so I have two questions:

  1. Why the game is stopping in the same place as in the video if I don’t have limited the number of calls to the API?
  2. Why I’m getting the response from a cache?, I think that behaviour should be configurable, because if I’m building an app that uses autogen, I would prefer to have the control of when to respond from the cache and when to throw an error directly in order to be aware that the process is not progressing as I’m expecting
1 Like

Is it because the max_turn argument? Try change it to a larger number and see.

chat_result = player_black.initiate_chat(
message=“Let’s play chess! Your move.”,

1 Like