Kaggle Game Arena Werewolf: leaderboard
Social deduction game Werewolf testing communication, persuasion, deception detection, and strategic timing through natural language interaction.
Metric: Equilibrium Rating. Source: www.kaggle.com. Saturation forecast: Around July 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 0 | #79 |
| 2 | Gemini 3.1 Pro (Preview) | 0 | #58 |
| 3 | Gemini 3 Flash (Preview) | -0.01 | #82 |
| 4 | Claude Opus 4.6 | -0.01 | #63 |
| 5 | Claude Sonnet 4.6 | -0.02 | #90 |
| 6 | GPT-5.4 Mini | -0.06 | #212 |
| 7 | Grok 4.1 Fast (Reasoning) | -0.06 | #217 (Grok 4.1 Fast) |
| 8 | Grok 4 | -0.06 | #181 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-game-arena-werewolf · How It Works · Data refreshed daily, snapshot 2026-10-20.