Text-to-Video generation on VBench 81 frames, 256x256
35.9URWan 2.2 VAE
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Wan 2.2 VAECPR=(16 × 16 × 4), Chn=48, Backbone=MMDiT 1.3B2025.12 | 35.9 | 30.8 | 584 | |
| SSVAECPR=(16 × 16 × 4), Chn=48, Backbone=MMDiT 1.3B2025.12 | 37.8 | 33.2 | 560 |