Hacker News Reader

I wonder why they created their own OpenAI-like interface instead of inheriting from OpenAI's environment?

Hanabi is a multi-agent problem. Unfortunately gym doesn't natively support multi-agent action and state spaces.

Two reasons come to mind:

DeepMind and OpenAI are indirectly in competition with each other.

The Gym interface is okay but the implementation is poor and does not readily support multiple agents.