PureDocBench - Clean: leaderboard
The clean track of PureDocBench (2026), a source-traceable document-parsing benchmark of 1,475 pages: pages rendered straight from their HTML/CSS source. The Overall score averages text accuracy, formula CDM and table TEDS on a 0 to 100 scale; higher is better.
Metric: Overall parsing score (0-100): mean of text accuracy (100 times one minus the text normalized edit distance), formula CDM and table TEDS, on the clean track (pages rendered straight from their HTML/CSS source) of 1,475 source pages in 10 domains and 66 subcategories, rendered from LLM-written HTML/CSS with source-linked annotations for text, formulas, tables and reading order; higher is better. Source: arxiv.org. Saturation forecast: Around December 2027. 58 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | TeleOCR | 87.16 | |
| 2 | GLM-5.3 Flash | 83.04 | #133 |
| 3 | Claude Opus 5 | 82.99 | #26 |
| 4 | OvisOCR2 | 81.53 | |
| 5 | Qwen 3.8 27B | 80.45 | #221 |
| 6 | Gemini 3.6 Flash | 79.97 | #81 |
| 7 | WeVisDoc-4B | 79.59 | |
| 8 | WeVisDoc-2B | 79.08 | |
| 9 | FD-RL | 78.38 | |
| 10 | ABot-OCR | 76.92 | |
| 11 | Logics-Parsing-v2 | 76.35 | |
| 12 | Dots.mocr | 76.27 | |
| 13 | Qwen 3.5 122B A10B | 76.14 | #346 |
| 14 | MinerU-2.5-Pro-1.2B | 75.87 | |
| 15 | Youtu-Parsing | 75.02 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=puredocbench-clean · How It Works · Data refreshed daily, snapshot 2026-11-21.