Real-world Multimodal Interaction on RealWorldQA (test)
77.8AccuracyQwen2-VL-72B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2-VL-72BModel Scale=72B2024.12 | 77.8 | |
| GPT-4o2024.12 | 76.5 | |
| LLaVA-OV-72BModel Scale=72B, Single-image variant (SI)=true2024.12 | 73.8 | |
| InternVL2-76BModel Scale=76B2024.12 | 72.7 | |
| LLaVA-OV-72BModel Scale=72B2024.12 | 71.9 | |
| MAmmoTH-VL-8BModel Scale=8B, Single-image variant (SI)=true2024.12 | 71.3 | |
| Molmo-7B-DModel Scale=7B2024.12 | 70.7 | |
| Gemini-1.5-Pro2024.12 | 70.4 | |
| Qwen2-VL-7B-InsModel Scale=7B2024.12 | 70.1 | |
| MAmmoTH-VL-8BModel Scale=8B2024.12 | 69.9 | |
| INXComp-2.5-7BModel Scale=2.5B2024.12 | 67.8 | |
| LLaVA-OV-7BModel Scale=7B2024.12 | 66.3 | |
| LLaVA-OV-7BModel Scale=7B, Single-image variant (SI)=true2024.12 | 65.5 | |
| MiniCPM-V-2.6-7BModel Scale=7B2024.12 | 65 | |
| InternVL-2-8BModel Scale=8B2024.12 | 64.4 | |
| Cambrian-1-8BModel Scale=8B2024.12 | 64.2 | |
| Llama-3.2-11B-Vision-InsModel Scale=11B2024.12 | 63.3 | |
| Claude-3.5-Sonnet2024.12 | 60.1 |