Image Question Answering on Prism (test)
86.04AI2D Score
Qwen3-VL-235B-Think
Updated 6d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.05 | 86.04 | 70.27 | 66.01 | 47.64 | 67.49 | |
2026.05 | 85.4 | 69 | 62.88 | 52.88 | 67.54 | |
2026.05 | 83.81 | 67.13 | 62.48 | 52.88 | 66.58 | |
2026.05 | 83.71 | 66.87 | 63.27 | 52.36 | 66.56 | |
2026.05 | 83.19 | 66.53 | 60.92 | 48.17 | 64.7 | |
2026.05 | 81.83 | 65.2 | 60.78 | 50.79 | 64.65 |