Video-to-audio generation on UnAV100
1.8FD (VGG)MMHNet - L
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| MMHNet - LSize=1.09B2026.02 | 1.8 | 5.29 | 209.06 | 1.49 | 8.1 | 7.35 | 36.27 | 0.41 | |
| MMHNet - SSize=157M2026.02 | 3.35 | 5.87 | 217 | 1.29 | 7.62 | 8.21 | 36.82 | 0.439 | |
| LoVASize=1.06B2026.02 | 3.36 | 7.5 | 223.29 | 1.56 | 6.17 | 8.46 | 24.62 | 1.232 | |
| MMAudio-LSize=1.03B2026.02 | 3.86 | 9.01 | 296.79 | 1.79 | 6.18 | 6.41 | 30.71 | 0.593 | |
| MMAudio-L + NTKSize=1.03B2026.02 | 4.08 | 8.43 | 301.45 | 1.76 | 6.05 | 6.17 | 30.24 | 0.599 | |
| V-AURASize=695M2026.02 | 4.57 | 6.16 | 321.64 | 1.69 | 7.35 | 5.22 | 29.36 | 1.191 | |
| MMAudio-SSize=157M2026.02 | 4.62 | 10.69 | 349.11 | 1.85 | 5.92 | 6.02 | 28.63 | 0.906 | |
| HunyuanVideo-Foley-XXLSize=5.13B2026.02 | 4.89 | 10.28 | 284.36 | 1.8 | 5.6 | 6.01 | 32.9 | 0.757 |