| mathematical-reasoning-on-mathverse | Qwen2.5-VL-32B + DIEM | #13 | Accuracy: 58 |
| multimodal-reasoning-on-mmmu | Qwen2.5-VL-32B + DIEM | #23 | Accuracy: 71.9 |
| vision-language-understanding-on-mmstar | Qwen2.5-VL-32B + DIEM | #37 | Accuracy: 71.2 |
| multimodal-reasoning-on-mathvision | Qwen2.5-VL-32B + DIEM | #39 | Accuracy: 40.3 |
| multimodal-reasoning-on-mathvista | Qwen2.5-VL-32B + DIEM | #40 | Accuracy: 76.9 |
| mathematical-reasoning-on-math-500 | Qwen3-4B + DIEM | #115 | Accuracy: 65.9 |
| mathematical-reasoning-on-aime24 | Qwen3-4B + DIEM | #128 | Acc: 16.2 |
| mathematical-reasoning-on-aime-2025 | Qwen3-4B + DIEM | #138 | Accuracy: 14.2 |