PureDocBench - Digital Degraded - Formula CDM: leaderboard

Formula recognition 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), scored by Character Detection Matching of rendered formulas from 0 to 100. Higher is better.

Metric: Formula Character Detection Matching score (0-100) between the rendered predicted and reference LaTeX formulas, 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; higher is better. Source: arxiv.org. Saturation forecast: Around November 2028. 58 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3.6 Flash78.88#84
2GLM-5.3 Flash78.51#137
3TeleOCR76.12
4Claude Opus 573.43#26
5Qwen3.8-Flash-Next72.87#122
6GPT-5.6 Sol72.83#36
7Qwen 3.8 27B71.98#232
8WeVisDoc-4B70.75
9WeVisDoc-2B70.34
10Qwen 3.5 4B68.88#1270
11Qwen 3.5 122B A10B67.82#350
12Qwen 3.6 35B A3B67.67#485
13Logics-Parsing-v267.33
14FD-RL67.16
15Qwen 3.5 9B67#838

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=puredocbench-digital-degraded-formula-cdm · How It Works · Data refreshed daily, snapshot 2026-11-22.