Great work on this project! I have a question regarding the current framework when running reinforcement learning training in OSWorld. Is it possible to use my own CPU server to run the Docker environment instead of relying on a cloud server?

[chatgptjzy-commits]

[yinjjiew]

I have tried to do so before. Sadly, it is not quite possible since the number of environments to parallelize in RL (64 in our setting) is too large for most personal CPU server.