SocialGrid (10x10 Rooms) - Impostor Win Rate: leaderboard
Metric: Impostor win rate (%) in a 30-matchup SocialGrid league in which one model plays all five crewmates and another both impostors in an Among Us-style gridworld (10x10 cells per room, 2x2 rooms, planning oracle on full assistance); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-OSS-120B | 98.5 |
| 2 | Llama 3.1 70B Instruct | 97.6 |
| 3 | Gemma 3 27B (IT) | 93.3 |
| 4 | DeepSeek R1 Distill Llama 70B | 92.7 |
Interactive version: theaggregate.ai/benchmark?slug=socialgrid-10x10-rooms-impostor-win-rate · How It Works · Data refreshed daily, snapshot 2026-10-07.