Parsing document structure and content
981 annotated PDF pages across 9 document categories. Tests end-to-end document parsing including text, tables, and formulas.
Leading models on OmniDocBench.
| # | Model | layout-map | Year | Source |
|---|---|---|---|---|
| ★ | MinerU 2.5 | 97.5 | 2025 | paper ↗ |
| 2 | GLM-OCR | 94.6 | 2026 | paper ↗ |
| 3 | GLM-OCR | 94.6 | 2026 | paper ↗ |
| 4 | PaddleOCR-VL-1.5 | 94.5 | 2026 | paper ↗ |
| 5 | PaddleOCR-VL | 93.5 | 2025 | paper ↗ |
| 6 | Qianfan-OCR | 93.1 | 2026 | paper ↗ |
| 7 | Qianfan-OCR | 93.1 | 2026 | paper ↗ |
| 8 | FireRed-OCR-2B | 92.9 | 2026 | paper ↗ |
| 9 | PaddleOCR-VL | 92.9 | 2025 | paper ↗ |
| 10 | PaddleOCR-VL 0.9B | 92.6 | 2025 | paper ↗ |
Questions, claims, and experiments linked through CodeSOTA’s public evidence graph.
Didn't find the model, metric, or dataset you needed? Tell us in one line. We read every message and reply within 48 hours.
3 datasets tracked for this task.
Still looking for something on Document Parsing? A missing model, a stale score, a benchmark we should cover — drop it here and we'll handle it.
Real humans read every message. We track what people are asking for and prioritize accordingly.