OpenExempt: leaderboard
Diagnostic legal benchmark for reasoning over U.S. Bankruptcy Code exemption scenarios.
Metric: Mean F1 (imported). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 72.31 | #124 |
| 2 | Gemini 2.5 Pro | 71.71 | #152 |
| 3 | GPT-5 | 71.67 | #98 |
| 4 | Gemini 2.5 Flash | 65.76 | #245 |
| 5 | DeepSeek R1 | 64.63 | #251 |
| 6 | O4 Mini | 59.35 | #180 |
| 7 | Claude Sonnet 4 | 57.37 | #199 |
| 8 | GPT-4.1 | 47.77 | #248 |
| 9 | DeepSeek V3 | 47.67 | #321 |
| 10 | Llama 4 Maverick | 40.92 | #464 |
| 11 | Claude 3.5 Haiku | 36.13 | #568 |
| 12 | Llama 4 Scout | 31.91 | #656 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openexempt · How It Works · Data refreshed daily, snapshot 2026-10-20.