oxwhirl/smac

SMAC: The StarCraft Multi-Agent Challenge

50
/ 100
Established

This project helps AI researchers and academics working on multi-agent reinforcement learning (MARL) to test and develop their algorithms. It provides a standardized environment using Blizzard's StarCraft II game, specifically focusing on scenarios where individual units are controlled by separate AI agents. Researchers can input their MARL algorithms and receive battle outcomes and agent behaviors within these complex real-time strategy settings.

1,330 stars. No commits in the last 6 months.

Use this if you are developing or evaluating cooperative multi-agent reinforcement learning algorithms and need a challenging, decentralized micromanagement testbed.

Not ideal if you are a StarCraft II player looking for AI opponents or a game developer wanting to integrate general AI into your game.

multi-agent AI research reinforcement learning game AI development AI algorithm testing real-time strategy game simulation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 24 / 25

How are scores calculated?

Stars

1,330

Forks

235

Language

Python

License

MIT

Last pushed

Feb 18, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/oxwhirl/smac"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.