OmniGameArena - ObstacleRun2D — leaderboard

Metric: Normalized game score (%). Source: arxiv.org. 10 models tracked.

Top models

#ModelScore
1GPT-5.547.3
2Claude Opus 4.633.8
3Gemini 3.1 Flash Lite (Preview)27.8
4Claude Opus 4.722
5GPT-5.412.2
6Qwen 3.5 397B A17B11.4
7Kimi K2.510.9
8Gemini 3.1 Pro (Preview)10.2
9Qwen 3.5 122B A10B9.2
10Claude Sonnet 4.67.5

Interactive version: aibenchmarks.dev/benchmark?slug=omnigamearena-obstaclerun2d · How the rankings work · Data refreshed daily, snapshot 2026-07-06.