Text-to-Video Generation on T2V-CompBench 55 (test)
0.417Motion ScoreCVG
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| CVGBackbone=Wan2.2-14B2026.05 | 0.417 | 0.535 | 0.667 | 0.888 | 0.117 | 0.795 | 0.742 | 0.594 | |
| CVGBackbone=CogVideoX-5B2026.05 | 0.412 | 0.398 | 0.624 | 0.786 | 0.037 | 0.663 | 0.622 | 0.506 | |
| TTOMBackbone=Wan2.2-14B2026.05 | 0.348 | 0.533 | 0.618 | 0.872 | 0.102 | 0.806 | 0.751 | 0.579 | |
| CVGBackbone=Wan2.2-5B2026.05 | 0.338 | 0.478 | 0.603 | 0.816 | 0.073 | 0.735 | 0.682 | 0.532 | |
| TTOMBackbone=CogVideoX-5B2026.05 | 0.332 | 0.392 | 0.566 | 0.742 | 0.029 | 0.671 | 0.638 | 0.482 | |
| Wan2.2-14BModel=Wan2.2-14B2026.05 | 0.301 | 0.521 | 0.592 | 0.857 | 0.091 | 0.792 | 0.736 | 0.556 | |
| TTOMBackbone=Wan2.2-5B2026.05 | 0.287 | 0.474 | 0.561 | 0.804 | 0.063 | 0.731 | 0.689 | 0.507 | |
| CogVideoX-5BModel=CogVideoX-5B2026.05 | 0.266 | 0.371 | 0.517 | 0.616 | 0.022 | 0.533 | 0.607 | 0.419 | |
| Wan2.2-5BModel=Wan2.2-5B2026.05 | 0.248 | 0.462 | 0.533 | 0.781 | 0.051 | 0.704 | 0.671 | 0.486 |