Skip to content

kaya-go/kaya

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

340 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Kaya Logo

Kaya

Play, study, and analyze Go — right from your browser or desktop.

Kaya is a free, open-source Go (Baduk/Weiqi) application with AI-powered analysis, board recognition from photos, and a beautiful modern interface. Available on Windows, macOS, Linux, and the Web.

Download Web App Next


Kaya — Go game with AI analysis

✨ Features

Game & Study

  • 🎯 Complete Go rules on 9×9, 13×13, and 19×19 boards
  • 🌳 Visual game tree with variation support
  • 📄 SGF import/export, drag & drop, OGS URL import
  • ✏️ Edit mode with stones, markers, labels, and annotations
  • 📚 Game library — organize games in folders
  • 🎯 Score estimation with interactive dead stone marking

AI & Analysis

  • 🤖 Live win rate, move suggestions, and full game analysis (KataGo via ONNX)
  • 📊 Analysis graph with move quality colors
  • 🗺️ Ownership heatmap for territory visualization

Board Recognition

  • 📷 Snap a photo of a real board → get a playable SGF
  • 🧠 Powered by Moku AI (RT-DETR model)
Kaya — Board recognition from photo with Moku AI
Board recognition: snap a photo of any Go board and import it instantly

More

  • 🎮 Keyboard shortcuts, gamepad support, mouse wheel navigation
  • 🎨 6 board themes + dark/light mode
  • 🌍 8 languages (EN, ZH, KO, JA, FR, DE, ES, IT)
  • 📱 Responsive on mobile, tablet, and desktop

Platform Support

  • 🖥️ Desktop — Native performance on Windows, macOS, and Linux via Tauri
  • 🌐 Web — Play directly in your browser (works on mobile and tablet too)
    • Stable — Latest official release
    • Next — Built from main branch (newest features)
  • 📱 PWA — Install the web app on any device for offline use (no app store needed)

🚀 Get Started

Platform How
🌐 Web Open kayago.app — nothing to install
🪟 Windows Download .exe
🍎 macOS Download .dmg
🐧 Linux Download .AppImage

🛠️ Tech Stack

Layer Technology
Frontend React 19 + TypeScript 5 + Rsbuild
Desktop Tauri v2 (Rust backend)
AI KataGo via ONNX Runtime
Board Recognition Moku AI (RT-DETR) + classic CV pipeline
Build Bun workspaces (monorepo, 14 packages)
Go logic TypeScript ports from Sabaki

GitHub Release GitHub Downloads CI License: AGPL v3


🤝 Contributing

Contributions are welcome — bug reports, feature ideas, or code!


📜 License

AGPL-3.0 © 2025-2026 Hadim


🙏 Acknowledgments

  • Sabaki — Core Go libraries and inspiration
  • Tauri — Lightweight desktop framework
  • KataGo — AI analysis engine

"Kaya" (榧) — the Japanese nutmeg tree, whose wood makes the finest Go boards.

Made with ❤️ for the Go community

⬆ Back to top

Sponsor this project

 

Packages

 
 
 

Contributors