Learning by building a multi-agent design lab. Yet, single agent works better

 (github.com)

4 points | by benoit_cotte 6 hours ago ago

3 comments