Video Generation on Human Video Generation Evaluation Set (test)
85.1Human Anatomy ScoreEchoMotion
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| EchoMotionBackbone=Wan-5B2025.12 | 85.1 | 99.3 | 64 | 58.3 | 81 | 81.5 | 81.6 | |
| Video TuningBackbone=Wan-5B2025.12 | 83.1 | 98.7 | 63.1 | 57.9 | 72.3 | 79.6 | 70.2 | |
| Wan-5BModel Size=5B, Variant=Base2025.12 | 83 | 98.9 | 62.2 | 58.3 | 72.8 | 78.9 | 68.9 | |
| EchoMotionBackbone=Wan-1.3B2025.12 | 79.6 | 98.9 | 61.9 | 60 | 71.3 | 73.2 | 66.1 | |
| Wan-1.3BModel Size=1.3B, Variant=Base2025.12 | 78.1 | 98.2 | 60.6 | 60.1 | 68.2 | 70.3 | 64 | |
| Video TuningBackbone=Wan-1.3B2025.12 | 77.4 | 98.3 | 61.6 | 59.7 | 69.3 | 73.2 | 65.5 | |
| CogVideoX1.5-5BModel Size=5B, Version=1.52025.12 | 65.3 | 98.5 | 54.4 | 53.2 | 62.5 | 60.4 | 59.4 | |
| CogVideoX-2BModel Size=2B2025.12 | 61.7 | 97 | 49.4 | 51.6 | 55.3 | 52.1 | 53.6 |