0ad4AI

A standard format for offline reinforcement learning datasets, with popular reference datasets and related utilities

Git mirror of the 0 A.D. source code (http://trac.wildfiregames.com/browser)

Code for Cicero, an AI agent that plays the game of Diplomacy with open-domain natural language negotiation.

A simple and highly efficient RTS-game-inspired environment for reinforcement learning (formerly Gym-MicroRTS)

A simple and highly efficient RTS-game-inspired environment for reinforcement learning

Sample script for using the RL interface in 0 AD from NodeJS

AI development for 0ad RTS game

A fast and simple framework for building and running distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyperparameter tuning library.

An RTS game built using Python and Pygame modules. Inspired by Google Deepmind's pysc2 interface.