Multimodal Reasoning and Mathematics on AI2D
85.6AccuracyDenseMLLM-4B
Evaluation results
| Method | Links | |
|---|---|---|
| DenseMLLM-4BModel Scale=4B2026.02 | 85.6 | |
| Qwen3-VL-4BModel Scale=4B2026.02 | 84.1 | |
| InternVL-3.5-4BModel Scale=4B2026.02 | 82.6 |
| Method | Links | |
|---|---|---|
| DenseMLLM-4BModel Scale=4B2026.02 | 85.6 | |
| Qwen3-VL-4BModel Scale=4B2026.02 | 84.1 | |
| InternVL-3.5-4BModel Scale=4B2026.02 | 82.6 |