Physical Reasoning on WorldModelBench
41.8General Score
Qwen9B-base
Updated 22d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.05 | 41.8 | 41.6 | 41.7 | 47.8 | |
2026.05 | 37.3 | 46.9 | 42.1 | 48.3 | |
2026.05 | 28.2 | 35 | 31.6 | 37.7 | |
2026.05 | 25.2 | 34.4 | 29.8 | 34.3 | |
2026.05 | 21.3 | 40.9 | 31.1 | 35.7 | |
2026.05 | 17.3 | 15.9 | 16.6 | -6.3 | |
2026.05 | 17.2 | 17.4 | 17.3 | -6.5 | |
2026.05 | 16.2 | 10 | 13.1 | 14.5 | |
2026.05 | 1 | 5.5 | 3.3 | -1.1 |