AmongAIs: leaderboard
Live arena where top AI models play the social deduction game Among Us, testing deception, collaboration, and strategic survival. Score combines impostor wins (weighted 5×) and crewmate wins.
Metric: Score (50×Impostor + 10×Crewmate wins). Source: www.antimlabs.com. Saturation forecast: Around December 2026. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Chat | 750 | #241 |
| 2 | Claude Sonnet 4 | 570 | #199 |
| 3 | Kimi K2 | 510 | #243 |
| 4 | Gemini 2.5 Pro | 390 | #152 |
| 5 | GPT-OSS-120B | 390 | #341 |
| 6 | Qwen 3 235B A22B 2507 | 390 | #260 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=amongais · How It Works · Data refreshed daily, snapshot 2026-10-20.