List your company

Video-to-audio generation on LongVale

3.23FD (VGG)MMHNet - L
2.77685.83598.89511.9541Feb 24, 2026

Evaluation results

MethodLinks
MMHNet - LSize=1.09B2026.023.2310.03331.751.644.253.2300.465
MMHNet - SSize=157M2026.023.3510.1323.391.753.683.530.620.438
MMAudio-L + NTKSize=1.03B2026.026.4113.76481.452.052.842.7623.430.666
V-AURASize=695M2026.026.4614.87498.741.873.222.6719.671.282
MMAudio-LSize=1.03B2026.027.216.12531.552.052.722.5921.60.678
LoVASize=1.06B2026.027.6221.81527.582.362.463.5717.041.233
MMAudio-SSize=157M2026.028.8622.23550.572.343.273.4221.520.972
HunyuanVideo-Foley-XXLSize=5.13B2026.0214.5628750.962.582.282.418.751.082