board-game-ai

🏆 Threes! AI: deck-aware expectimax, N-tuple TD learning, and AlphaZero

AlphaZero-style Quoridor engine trained from zero knowledge — no human games, no heuristics. 24 h of self-play on one consumer GPU, after which it beats the previous strongest public Quoridor AI with no search at all. Play it in your browser.

Python framework for building Hex Connect-6 AI bots. AlphaZero self-play, MCTS, neural networks, C game engine, live training dashboard. Home of the Orca bot.

A four-player chess engine built to play the strongest 4PC it can. NNUE evaluation, an alpha-beta core in C and a Python root. Perft-verified move generation, all five starting setups, and a browser board to play against it. Plays Teams; FFA rules implemented, FFA search in progress.

A Go board game engine for small boards. It plays with Monte Carlo tree search and speaks the Go Text Protocol.

Open-source reinforcement-learning engine for knight-move Gomoku on a 15x15 board.

Chinese Checkers AI engine, JumpStar self-play model, and CCERL benchmark artifacts

Open-source Renju Taraguchi-10 training and review lab powered by Rapfi

Competitive Othello (Reversi) AI using alpha-beta pruning, Zobrist hashing, and iterative deepening. Playable as Player vs Player, Player vs Agent, or Agent vs Agent.

End-to-end AlphaZero reinforcement learning engine for the board game Snort featuring PyTorch Dual-Head ResNet, PUCT Monte Carlo Tree Search, self-play training loop, automated Elo tournament, and an interactive Flask Web GUI.

Kōnane is a traditional Hawaiian strategy board game.

Open-source Algerian/Spanish checkers engine and WPF application featuring alpha-beta AI, benchmarking, and bitboard/FPGA-oriented research.

A Discord bot that plays Magic: The Gathering — Claude-vs-Claude autoplay or play it yourself, with a tiered rules engine (templates → SpellResolver → optional XMage bridge → LLM judge).