Rare find

SnakeBench. Watch AI models play snake against each other and see who wins.

Problem

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Audience

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Market

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Pricing

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

June 2026

Pause automated new-model evaluation in cron service Jun 10
Adjust default video crop to 0.16 (keep middle 68%) Jun 5
Pause auto-activation of new models to official providers only Jun 5

May 2026

Install ffmpeg in Celery worker image for video encoding May 30
Bump playwright to 1.50.0 for Ubuntu 24.04 (noble) compatibility May 30
Switch replay video generation to Playwright capture of live frontend May 30
Add frontier-tier prior and adaptive game budget at top of ladder May 29
Replace placement with noisy binary search (Bayesian, 10-game budget) May 29
Fix duplicate-opponent dispatches by reserving DB row at enqueue time May 29

March 2026

Fix eval batch to finalize completed models even with pending games Mar 8

February 2026

Disable ladder matchmaking to reduce recurring game costs Feb 24
Fix FLASK_URL env var mismatch across frontend Feb 24
Upgrade Railway frontend to Node.js 20 for Supabase compatibility Feb 24
Add .env* to .gitignore and untrack .env.local Feb 24
Bump PyJWT to 2.10.1 to fix supabase-auth dependency conflict Feb 24
qol Feb 24
Merge pull request #14 from gkamradt/claude/add-latest-games-nav-Ys6Ro Feb 1
Update package-lock.json from npm install Feb 1
Fix build issues: use dynamic rendering and CSS font loading Feb 1
Add Latest Games navigation and page Feb 1