InduOCRBench - Formula: leaderboard
The formula recognition score of InduOCRBench (2026), a document-parsing benchmark of 570 real industrial documents in 11 challenge types: Character Detection Matching between rendered predicted and reference formulas, 0 to 100. Higher is better.
Metric: Formula recognition score (0-100): Character Detection Matching (CDM) between the rendered predicted and reference formulas on 570 industrial documents (3,402 pages) in 11 challenge types (complex background, high pixel count, ultra-long, multi-column, ultra-wide, historical books, handwriting, multi-font, visual style, watermark, cross-page tables) plus normal documents, each OCR method converting pages to structured Markdown (general VLMs with a standardized prompt, pipeline tools at default configuration, closed solutions through their official APIs); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 77.2 | #276 |
| 2 | MinerU2.5-2509-1.2B | 75.4 | |
| 3 | Qwen 3 VL 235B A22B Instruct | 74.8 | #532 |
| 4 | PaddleOCR-VL | 74.6 | |
| 5 | HunyuanOCR | 65.6 | |
| 6 | Mineru2-pipeline | 63.2 | |
| 7 | DeepSeek-OCR | 61.8 | |
| 8 | GPT-4o | 58.1 | #751 |
| 9 | Doc2X | 56.3 | |
| 10 | PP-StructureV3 | 53.7 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=induocrbench-formula · How It Works · Data refreshed daily, snapshot 2026-11-21.