OmniGameArena - SharedFloor: leaderboard

The "SharedFloor" subtask of OmniGameArena. OmniGameArena (arXiv) — Unreal-Engine-5 game environments testing vision-language agents on real-time perception, control and planning across solo and co-op games; normalized game score (%).

Metric: Normalized game score (%). Source: arxiv.org. Saturation forecast: Around January 2027. 10 models tracked.

Top models

#ModelScoreOverall rank
1GPT-5.536.8#31
2Gemini 3.1 Pro (Preview)33.6#58
3Claude Opus 4.615.2#63
4Claude Sonnet 4.614.8#90
5Claude Opus 4.713.6#49
6Kimi K2.57.2#144
7GPT-5.46.8#79
8Gemini 3.1 Flash Lite (Preview)5.2#194
9Qwen 3.5 122B A10B0.8#173
10Qwen 3.5 397B A17B0#146

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=omnigamearena-sharedfloor · How It Works · Data refreshed daily, snapshot 2026-10-20.