SnakeBench — leaderboard

LLM Battle Snake arena where AI models play the classic Snake game competitively, testing strategic planning and decision-making.

Metric: TrueSkill Rating. Source: snakebench.com. Status: saturation imminent. 318 models tracked.

Top models

#ModelScore
1Gemini 3 Pro (Preview)38
2GPT-5.537.4
3Gemini 3.1 Pro (Preview)37
4GPT-5.1 Codex Mini36.5
5O336.4
6GPT-5.1 Codex Max36.4
7Claude Opus 4.835.8
8GPT-535
9Qwen 3.7 Max34.1
10O4 Mini33.9
11O4 Mini (High)33.7
12GPT-5 Codex33.3
13GPT-5.233.2
14Gemini 2.0 Flash33.2
15Gemini 3.5 Flash33.2

Interactive version: theaggregate.ai/benchmark?slug=snakebench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.