PureDocBench - Clean - Formula CDM: leaderboard
Formula recognition on the clean track of PureDocBench (2026), a document-parsing benchmark of 1,475 source-traceable pages (pages rendered straight from their HTML/CSS source), 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 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 2028. 58 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | TeleOCR | 82.33 | |
| 2 | GLM-5.3 Flash | 80.16 | #137 |
| 3 | Gemini 3.6 Flash | 78.29 | #84 |
| 4 | Claude Opus 5 | 75.92 | #26 |
| 5 | GPT-5.6 Sol | 74.96 | #36 |
| 6 | Qwen3.8-Flash-Next | 72.28 | #122 |
| 7 | Qwen 3.8 27B | 71.68 | #232 |
| 8 | WeVisDoc-2B | 71.25 | |
| 9 | Qwen 3.6 35B A3B | 70.39 | #485 |
| 10 | Qwen 3.5 4B | 69.96 | #1270 |
| 11 | WeVisDoc-4B | 69.95 | |
| 12 | OvisOCR2 | 69.66 | |
| 13 | Qwen 3.6 27B | 69.36 | #358 |
| 14 | FD-RL | 68.21 | |
| 15 | Qwen 3.5 122B A10B | 67.96 | #350 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=puredocbench-clean-formula-cdm · How It Works · Data refreshed daily, snapshot 2026-11-22.