RevengeBench - Poker: leaderboard
RevengeBench asks an agent to reconstruct a hidden game bot's policy from observed behavior and experiments. This series reports Poker distance reduction relative to a starter policy, in percent, under mini-SWE-agent with the best of five rounds reported. It measures how closely the recovered policy matches held-out actions; higher is better and negative values mean worse than the starter.
Metric: Distance reduction (%; best round, mini-SWE-agent). Source: revengebench.com. Saturation forecast: Around June 2027. 12 models tracked. Updated 2 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 82.2 | #50 |
| 2 | GLM-5.1 | 78 | #195 |
| 3 | GPT-5 | 76.3 | #177 |
| 4 | Qwen 3.6 35B A3B | 74.4 | #484 |
| 5 | DeepSeek V4 Flash | 73.6 | #265 |
| 6 | DeepSeek V4 Pro | 73 | #198 |
| 7 | Gemma 4 31B | 68.5 | #385 |
| 8 | Gemma 4 26B A4B | 63.6 | #581 |
| 9 | Grok 4.1 Fast | 60.6 | #429 |
| 10 | Kimi K2.6 | 58.4 | #191 |
| 11 | GPT-5.4 Mini | 44.2 | #401 |
| 12 | GPT-OSS-120B | 40.5 | #707 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=revengebench-poker · How It Works · Data refreshed daily, snapshot 2026-11-17.