title: Faïence
emoji: 🀄
colorFrom: blue
colorTo: yellow
sdk: static
pinned: true
thumbnail: https://remifabre-faience.static.hf.space/social.png
short_description: Azul rules vs a self-play neural net, all in your browser
Faïence is a free, open-source implementation of the rules of Azul, the tile-laying game by Michael Kiesling: a fan project for research, with its own code and artwork, not affiliated with or endorsed by the game's publishers.
Your opponent is a neural network that learned the game from scratch by self-play. The net and its tree search run entirely in your browser: your browser downloads the model once and everything after that is local. No server plays for it, no account, no ads, nothing to buy.
It began as a machine-learning experiment: Ludometer, a research framework that measures how good a board game is from the shape of an AI's learning curve. The games people play here are the research material: when a game ends, the page sends an anonymous record (moves, deals, net, score; sharing can be switched off in Settings) to the public dataset RemiFabre/faience-games, where it becomes training data.
Code, training logs and methodology: RemiFabre/ludometer (MIT).