murongg

murongg / ai-zhuqi-battle

Public

一个LLM与LLM之间的五子棋/象棋对战小游戏

ai-zhuqi-battle.vercel.app
15
4
100% credibility
Found Mar 05, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A web app for watching two AI services battle in real-time games of Gomoku, Chinese Chess, or International Chess, with move thoughts and stats.

How It Works

1
🔍 Find the fun AI game battles

You stumble upon this cool project online where AIs play board games against each other, like a never-ending chess match between smart bots.

2
💻 Get it running on your computer

You download and start the app on your own machine, and a colorful game screen pops up ready for action.

3
🎮 Pick your favorite game

Choose between five-in-a-row, Chinese chess, or international chess, and the board appears looking just like the real thing.

4
🧠 Connect smart brains for each side

For each player side, you link up an AI thinking service by filling in a few simple details, so both sides can plan their moves.

5
▶️ Hit start and watch the battle

Click the big button to begin, and the AIs take turns making moves while sharing their step-by-step thoughts right on screen.

🏆 Enjoy the wins, draws, and replays

See who wins, track the score over many games, restart anytime, and feel the thrill of AI showdowns whenever you want.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ai-zhuqi-battle?

This TypeScript Next.js app pits two LLMs head-to-head in zhuqi battles across Gomoku (15x15), Xiangqi (Chinese chess, 9x10), and international chess (8x8). Configure separate API endpoints, models, and keys for each side via side panels—black vs. white or red vs. black—then hit "start battle" to watch them alternate turns with adjustable delays. It solves the hassle of building game engines by handling rules, visuals, win detection, and LLM prompting on the frontend, proxying calls through a simple API route.

Why is it gaining traction?

Seamless multi-provider support (OpenAI-compatible endpoints) with persistent conversation context per side keeps games coherent, while fallback moves prevent stalls from bad LLM outputs. Canvas-rendered boards show last moves highlighted, recent history, AI "thoughts," and win stats—zero setup beyond pasting keys. The game-switcher and localStorage configs make it addictive for quick battles without redeploying.

Who should use this?

AI prompt engineers benchmarking LLM strategy in zero-sum games. Hobbyists pitting cheap models like GPT-4o-mini against premium ones in Xiangqi or chess. TypeScript devs prototyping LLM-driven game agents without chess engine libs.

Verdict

Fun, playable demo at 15 stars and 1.0% credibility—early but stable TypeScript with solid README for forking. Skip for production; grab for LLM battle experiments or extending prompts.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.