QueensGambit/CrazyAra

A Deep Learning UCI-Chess Variant Engine written in C++ & Python :parrot:

55
/ 100
Established

This project provides a powerful chess engine that can play various chess variants using advanced deep learning. It takes in game positions for variants like Crazyhouse, Chess960, King of the Hill, and Three-Check, and outputs optimal moves. This is for chess enthusiasts, variant players, or researchers interested in advanced AI for strategy games.

283 stars.

Use this if you want to play against, analyze, or integrate a strong AI opponent for various chess variants into a chess GUI.

Not ideal if you are looking for a standard chess engine or a simple tool for beginners, as it focuses on advanced AI for variants and requires a separate chess GUI.

chess-variants board-games game-AI strategy-games competitive-gaming
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

283

Forks

44

Language

Jupyter Notebook

License

GPL-3.0

Last pushed

Mar 02, 2026

Commits (30d)

0

Get this data via API

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

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