Visual Reasoning on HR-Bench 4K FSP
96.5ACC
RTWI
Updated 3d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.02 | 96.5 | 66.2 | |
2026.02 | 96 | 60.3 | |
2026.02 | 95.8 | 59.8 | |
2026.02 | 95.5 | 57.7 | |
2026.02 | 95.3 | - | |
2026.02 | 95.3 | 60.7 | |
2026.02 | 95.3 | 62.5 | |
2026.02 | 93.8 | - | |
2026.02 | 91.3 | - | |
2026.02 | 91.3 | 61.9 | |
2026.02 | 91 | - | |
2026.02 | 91 | - | |
2026.02 | 90.5 | 57.3 | |
2026.02 | 90 | 64.5 | |
2026.02 | 89.8 | 56.4 | |
2026.02 | 89.5 | - | |
2026.02 | 89.5 | 58.9 | |
2026.02 | 89.5 | 55.8 | |
2026.02 | 89.5 | 53.9 | |
2026.02 | 89.5 | 61.9 | |
2026.02 | 89.5 | - | |
2026.02 | 89.5 | 51.9 | |
2026.02 | 89.5 | 40.5 | |
2026.02 | 89.5 | 47.6 | |
2026.02 | 88 | 32.3 | |
2026.02 | 87.8 | - | |
2026.02 | 83 | - | |
2026.02 | 66.8 | - | |
2026.02 | 66.8 | - |