Audio-Visual Generation on Verse-Bench (All subsets)
4.269IS (Score)MOVA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MOVAResolution=360p, s_B=1.0, dual CFG=false2026.02 | 4.269 | 0.475 | 0.286 | |
| MOVAResolution=360p, s_B=3.5, dual CFG=true2026.02 | 4.169 | 0.351 | 0.315 | |
| WAN2.1 + MMAudioResolution=720p2026.02 | 4.036 | 0.26 | 0.317 | |
| MOVAResolution=720p, s_B=1.0, dual CFG=false2026.02 | 3.936 | 0.485 | 0.277 | |
| MOVAResolution=720p, s_B=3.5, dual CFG=true2026.02 | 3.814 | 0.37 | 0.297 | |
| OviResolution=720p2026.02 | 3.68 | 0.515 | 0.19 | |
| LTX-2Resolution=720p2026.02 | 3.066 | 0.451 | 0.213 |