OCR Text Recognition on OmniDocBench
0.026NED (ALL avg)Qwen3-VL
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3-VLModel Type=Vision Language Models, Params=235B2026.03 | 0.026 | 0.016 | 0.026 | 0.034 | 0.023 | 0.023 | 0.03 | 0.023 | 0.046 | 0.029 | 0.17 | |
| PP-OCRv5Model Type=Pipeline Tools & Expert Vision Models, Params=5M2026.03 | 0.067 | 0.058 | 0.076 | 0.103 | 0.07 | 0.081 | 0.095 | 0.072 | 0.012 | 0.058 | 0.139 | |
| GOT-OCRModel Type=Pipeline Tools & Expert Vision Models, Params=0.58B2026.03 | 0.077 | 0.041 | 0.112 | 0.135 | 0.092 | 0.052 | 0.155 | 0.091 | 0.562 | 0.097 | 0.966 | |
| SuryaModel Type=Pipeline Tools & Expert Vision Models, Params=-2026.03 | 0.09 | 0.057 | 0.123 | 0.164 | 0.093 | 0.186 | 0.235 | 0.104 | 0.634 | 0.255 | 0.767 | |
| InternVL2Model Type=Vision Language Models, Params=76B2026.03 | 0.115 | 0.074 | 0.155 | 0.242 | 0.113 | 0.352 | 0.269 | 0.132 | 0.61 | 0.595 | 0.907 | |
| GPT4oModel Type=Vision Language Models, Params=-2026.03 | 0.122 | 0.02 | 0.224 | 0.125 | 0.167 | 0.14 | 0.22 | 0.168 | 0.115 | 0.132 | 0.718 | |
| MathpixModel Type=Pipeline Tools & Expert Vision Models, Params=-2026.03 | 0.137 | 0.033 | 0.24 | 0.261 | 0.185 | 0.121 | 0.166 | 0.18 | 0.038 | 0.638 | 0.185 | |
| Qwen2-VLModel Type=Vision Language Models, Params=72B2026.03 | 0.173 | 0.072 | 0.274 | 0.286 | 0.234 | 0.155 | 0.148 | 0.223 | 0.273 | 0.067 | 0.721 | |
| Tesseract-OCRModel Type=Pipeline Tools & Expert Vision Models, Params=-2026.03 | 0.324 | 0.096 | 0.551 | 0.25 | 0.439 | 0.328 | 0.331 | 0.426 | 0.117 | 0.984 | 0.969 | |
| EasyOCRModel Type=Pipeline Tools & Expert Vision Models, Params=-2026.03 | 0.329 | 0.26 | 0.398 | 0.445 | 0.366 | 0.287 | 0.388 | 0.36 | 0.97 | 0.926 | 0.997 |