General Visual Question Answering on MMBench 2024c (dev)
83.3AccuracyInternVL2.5-MPO
Evaluation results
| Method | Links | |
|---|---|---|
| InternVL2.5-MPO2025.01 | 83.3 | |
| InternVL2.52025.01 | 83.1 | |
| InternVL22025.01 | 80.3 | |
| SAIL-VL-8B2025.01 | 79.5 | |
| Qwen2-VL2025.01 | 79.5 | |
| DeepSeekVL-2-Small2025.01 | 78.1 | |
| Aquila2025.01 | 74.4 | |
| InternVL2.52025.01 | 73.1 | |
| InternVL2.5-MPOTraining Strategy=MPO2025.01 | 72.5 | |
| SAIL-VLModel Size=2B2025.01 | 72.4 | |
| InternVL22025.01 | 70.3 | |
| Qwen2-VL2025.01 | 68.8 | |
| DeepSeekVL-2-TinyModel Size=Tiny2025.01 | 68.3 |