OmniGameArena - HandoffRun: leaderboard
The "HandoffRun" 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 February 2028. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 18.4 | #31 |
| 2 | Gemini 3.1 Pro (Preview) | 13.6 | #58 |
| 3 | Claude Opus 4.6 | 6.4 | #63 |
| 4 | GPT-5.4 | 6 | #79 |
| 5 | Claude Opus 4.7 | 4 | #49 |
| 6 | Gemini 3.1 Flash Lite (Preview) | 2 | #194 |
| 7 | Claude Sonnet 4.6 | 0.8 | #90 |
| 8 | Kimi K2.5 | 0.8 | #144 |
| 9 | Qwen 3.5 397B A17B | 0 | #146 |
| 10 | Qwen 3.5 122B A10B | 0 | #173 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=omnigamearena-handoffrun · How It Works · Data refreshed daily, snapshot 2026-10-20.