DGEval - Regulatory Recall: leaderboard
Metric: Accuracy (%). Source: arxiv.org. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 36.8 |
| 2 | GPT-5.5 | 30.5 |
| 3 | Gemini 3.1 Flash Lite | 22 |
| 4 | Claude Opus 4.7 | 20.2 |
| 5 | GPT-5.4 Mini | 15.6 |
| 6 | Claude Sonnet 4.6 | 11.4 |
| 7 | Llama 3.3 70B | 6.6 |
| 8 | Gemma 4 31B | 5.2 |
| 9 | Qwen 3.6 35B A3B | 2.1 |
| 10 | Qwen 3.5 9B | 1 |
| 11 | Llama 3.2 3B | 0.3 |
Interactive version: theaggregate.ai/benchmark?slug=dgeval-regulatory-recall · How It Works · Data refreshed daily, snapshot 2026-09-05.