PureDocBench - Digital Degraded - Reading Order Edit Distance: leaderboard

Reading-order normalized edit distance on the digital degraded track of PureDocBench (2026), a document-parsing benchmark of 1,475 source-traceable pages (pages under simulated aging, binding, photocopy, compression and lighting degradations). Percent; lower is better.

Metric: Reading-order normalized edit distance (%) between the predicted and reference block sequences, times 100, on the digitally degraded track (the same pages under ten simulated scene templates such as aging, binding, photocopying, compression, lighting and geometric artifacts) 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; lower is better. Source: arxiv.org. Saturation forecast: Not forecast. 58 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1TeleOCR23.7
2Dots.mocr30.9
3OvisOCR231.6
4Qwen 3.8 27B32.5#221
5FD-RL35
6WeVisDoc-4B35.6
7Qwen 3.5 122B A10B36.6#346
8Qwen 3.5 27B36.7#375
9MinerU2.5-2509-1.2B37.4
10WeVisDoc-2B37.4
11Logics-Parsing-v237.5
12Qwen 3.5 397B A17B37.6#264
13olmOCR-237.8
14MinerU-2.5-Pro-1.2B37.8
15Qwen 3.5 35B A3B37.9#492

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=puredocbench-digital-degraded-reading-order-edit-distance · How It Works · Data refreshed daily, snapshot 2026-11-21.