Pose-to-Video Generation on VACE-Benchmark
66.71Aesthetic QualityVDOT
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| VDOTBase Model=Wan-14B, #NFE=42025.12 | 66.71 | 94.99 | 80 | 70.4 | 98.45 | 94.26 | 84.13 | |
| VACEBase Model=Wan-14B, #NFE=1002025.12 | 66.69 | 95.32 | 75 | 68.48 | 98.56 | 94.19 | 83.04 | |
| ControlVideo2026.03 | 65.4 | 94.6 | 25 | 65.3 | 97.3 | 92.8 | 73.4 | |
| AVControl2026.03 | 63.6 | 93.1 | 84.2 | 68.5 | 98.9 | 94 | 83.7 | |
| VACE2026.03 | 60.2 | 94.9 | 75 | 64.7 | 98.6 | 94.8 | 81.4 | |
| VACEBase Model=LTX-0.9B, #NFE=802025.12 | 59.97 | 94.8 | 85 | 66.85 | 98.93 | 94.91 | 83.41 | |
| Text2Video-Zero2026.03 | 57.6 | 87.7 | 100 | 70.7 | 79.7 | 84.8 | 80.1 | |
| Follow-Your-Pose2026.03 | 48.8 | 86.8 | 100 | 67.4 | 90.1 | 80.2 | 78.9 |