RefereeBench - Foul and Penalty Reasoning: leaderboard
Metric: Accuracy (%) on foul and penalty reasoning questions (mean of Q3 and Q5), averaged across 11 sports; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 53.1 |
| 2 | Gemini 3 Pro | 53 |
| 3 | Seed 1.8 | 50.7 |
| 4 | GPT-5 | 47.8 |
| 5 | Claude Sonnet 4.5 | 42.5 |
| 6 | GPT-4o | 39.1 |
| 7 | Claude Haiku 4.5 | 38.7 |
Interactive version: theaggregate.ai/benchmark?slug=refereebench-foul-and-penalty-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-07.