# OmniDocBench

Tests how well AI can read a PDF page and turn it into clean text, including tables and math formulas. It matters for turning scanned or complex documents into usable data.

Category: ocr. Type: leaderboard. Status: active.
Metric: Overall document-parsing score (0-100, higher is better), averaging text accuracy, table accuracy and formula accuracy (higher is better).
Source: OpenDataLab / Shanghai AI Lab (OmniDocBench repo) — https://github.com/opendatalab/OmniDocBench, results as of 2026-11-12, retrieved 2026-11-17, license: not stated (dataset research-only, non-commercial per repo copyright statement).

| Rank | Model | Method type | Size | Overall | Text edit distance (lower better) | Formula CDM | Table TEDS |
|---|---|---|---|---|---|---|---|
| 1 | OXR-1.0 | Specialized VLMs | 1.3B | 97.03 | 0.0244 | 96.93 | 96.61 |
| 2 | TeleOCR | Specialized VLMs | 1.2B | 96.91 | 0.0267 | 96.5895 | 96.8183 |
| 3 | OvisOCR2 | Specialized VLMs | 0.8B | 96.47 | 0.0265 | 97.4854 | 94.5842 |
| 4 | MiniCPM-OCR 1.0 | Specialized VLMs | 1.3B | 96.44 | 0.0316 | 98.0814 | 94.3999 |
| 5 | PaddleOCR-VL-1.6 | Specialized VLMs | 0.9B | 96.34 | 0.0326 | 97.5304 | 94.7619 |
| 6 | MinerU2.5-Pro | Specialized VLMs | 1.2B | 95.75 | 0.036 | 97.45 | 93.42 |
| 7 | GLM-OCR | Specialized VLMs | 0.9B | 95.22 | 0.044 | 97.18 | 92.83 |
| 8 | PaddleOCR-VL-1.5 | Specialized VLMs | 0.9B | 94.93 | 0.038 | 96.89 | 91.67 |
| 9 | PaddleOCR-VL | Specialized VLMs | 0.9B | 94.18 | 0.040 | 95.91 | 90.65 |
| 10 | Unlimited-OCR | Specialized VLMs | 3B | 94.00 | 0.0394 | 95.7234 | 90.21 |
| 11 | Youtu-Parsing | Specialized VLMs | 2.5B | 93.74 | 0.044 | 93.63 | 92.02 |
| 12 | Qianfan-OCR | Specialized VLMs | 4B | 93.90 | 0.04 | 95.08 | 90.53 |
| 13 | Ovis2.6-30B-A3B | General VLMs | 30B | 93.70 | 0.035 | 95.17 | 89.44 |
| 14 | Logics-Parsing-v2 | Specialized VLMs | 4B | 93.33 | 0.041 | 95.65 | 88.42 |
| 15 | ABot-OCR | Specialized VLMs | 2B | 93.30 | 0.037 | 94.86 | 88.69 |
| 16 | FireRed-OCR | Specialized VLMs | 2B | 93.26 | 0.037 | 95.44 | 88.04 |
| 17 | MinerU-2.5 | Specialized VLMs | 1.2B | 93.04 | 0.045 | 95.77 | 87.88 |
| 18 | Gemini 3 Pro | General VLMs | - | 92.91 | 0.064 | 95.99 | 89.15 |
| 19 | Gemini 3 Flash | General VLMs | - | 92.62 | 0.066 | 95.16 | 89.29 |
| 20 | MonkeyOCR V2 | Specialized VLMs | - | 91.80 | 0.0636 | 93.4985 | 88.2628 |
| 21 | dots.ocr | Specialized VLMs | 3B | 90.77 | 0.048 | 89.95 | 87.18 |
| 22 | OpenDoc-0.1B | Specialized VLMs | 0.1B | 90.67 | 0.049 | 93.02 | 83.88 |
| 23 | DeepSeek-OCR 2 | Specialized VLMs | 3B | 90.25 | 0.050 | 91.84 | 83.89 |
| 24 | HunyuanOCR | Specialized VLMs | 1B | 89.95 | 0.088 | 87.68 | 91.01 |
| 25 | Qwen3-VL-235B | General VLMs | 235B | 89.78 | 0.063 | 92.55 | 83.07 |

Notes: End-to-end table, benchmark version v1.6_full, from repo README in the order listed (best first), top 25. Scores are mostly self-reported by model authors or the benchmark team.
