DocHop: leaderboard
Metric: Accuracy (%). Source: arxiv.org. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 62.83 |
| 2 | Gemini 2.5 Pro (Thinking) | 40.6 |
| 3 | Gemini 2.5 Flash (Thinking) | 32.02 |
| 4 | GPT-5 Mini | 28.25 |
| 5 | Gemini 2.5 Flash | 24.88 |
| 6 | Claude Sonnet 4.5 | 17.94 |
| 7 | Claude Haiku 4.5 | 16.35 |
Interactive version: theaggregate.ai/benchmark?slug=dochop · How It Works · Data refreshed daily, snapshot 2026-09-20.