OCR and Chart Understanding on TextVQA
87.4AccuracyQwen2.5-VL-7B +PC
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2.5-VL-7B +PCBackbone Model=Qwen2.5-VL-7B, Training Paradigm (+PC)=Yes2026.03 | 87.4 | |
| Qwen2.5-VL-7BBackbone Model=Qwen2.5-VL-7B, Training Paradigm (+PC)=No2026.03 | 84.9 | |
| Qwen2.5-VL-3B +PCBackbone Model=Qwen2.5-VL-3B, Training Paradigm (+PC)=Yes2026.03 | 84.8 | |
| MiMo-VL-7B +PCBackbone Model=MiMo-VL-7B, Training Paradigm (+PC)=Yes2026.03 | 84.3 | |
| MiMo-VL-7BBackbone Model=MiMo-VL-7B, Training Paradigm (+PC)=No2026.03 | 81.2 | |
| Qwen3-VL-4BModel Scale=4B2026.02 | 80.8 | |
| DenseMLLM-4BModel Scale=4B2026.02 | 79.6 | |
| Qwen2.5-VL-3BBackbone Model=Qwen2.5-VL-3B, Training Paradigm (+PC)=No2026.03 | 79.3 | |
| InternVL-3.5-4BModel Scale=4B2026.02 | 77.9 | |
| AVLM-2BModel Scale=2B2026.06 | 75.52 | |
| Gemma-4-E2B-itModel Scale=4B2026.06 | 61.19 | |
| Qwen2.5-Omni-3BModel Scale=3B2026.06 | 20.17 |