Document Understanding on InfoVQA (test)
84.5AccuracyQwen2-VL-72B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2-VL-72BModel Scale=72B2024.12 | 84.5 | |
| InternVL2-76BModel Scale=76B2024.12 | 82 | |
| Gemini-1.5-Pro2024.12 | 81 | |
| GPT-4o2024.12 | 79.2 | |
| Qwen2-VL-7B-InsModel Scale=7B2024.12 | 76.5 | |
| LLaVA-OV-72BModel Scale=72B2024.12 | 74.9 | |
| InternVL-2-8BModel Scale=8B2024.12 | 74.8 | |
| MAmmoTH-VL-8BModel Scale=8B, Single-image variant (SI)=true2024.12 | 74.8 | |
| LLaVA-OV-72BModel Scale=72B, Single-image variant (SI)=true2024.12 | 74.6 | |
| MAmmoTH-VL-8BModel Scale=8B2024.12 | 73.1 | |
| Molmo-7B-DModel Scale=7B2024.12 | 72.6 | |
| INXComp-2.5-7BModel Scale=2.5B2024.12 | 70 | |
| LLaVA-OV-7BModel Scale=7B2024.12 | 68.8 | |
| LLaVA-OV-7BModel Scale=7B, Single-image variant (SI)=true2024.12 | 65.3 | |
| Llama-3.2-11B-Vision-InsModel Scale=11B2024.12 | 65 | |
| Claude-3.5-Sonnet2024.12 | 49.7 | |
| Llava-CoT-11BModel Scale=11B2024.12 | 44.8 | |
| Cambrian-1-8BModel Scale=8B2024.12 | 41.6 |