MDPBench: leaderboard
Multilingual document parsing benchmark covering digital and photographed documents across scripts, layouts, and languages.
Metric: Overall Score (%). Source: huggingface.co. Saturation forecast: Around January 2027. 31 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 86.4 | #66 |
| 2 | Kimi K2.5 | 77.5 | #144 |
| 3 | Claude Sonnet 4.6 | 73.1 | #90 |
| 4 | Qwen 3 VL 8B Instruct | 68.3 | #420 |
| 5 | Qwen 3.5 9B | 65.7 | #379 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mdpbench · How It Works · Data refreshed daily, snapshot 2026-10-20.