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

#ModelScoreOverall rank
1TeleOCR87.16
2GLM-5.3 Flash83.04#133
3Claude Opus 582.99#26
4OvisOCR281.53
5Qwen 3.8 27B80.45#221
6Gemini 3.6 Flash79.97#81
7WeVisDoc-4B79.59
8WeVisDoc-2B79.08
9FD-RL78.38
10ABot-OCR76.92
11Logics-Parsing-v276.35
12Dots.mocr76.27
13Qwen 3.5 122B A10B76.14#346
14MinerU-2.5-Pro-1.2B75.87
15Youtu-Parsing75.02

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.