- Models
- ByteDance
- LatentSync
- LatentSync 1.6

LatentSync 1.6
latentsync-1.6
Audio-conditioned latent diffusion lip-sync model. Version 1.6 processes 512x512 video; requires source video and target audio, with official inference code and pretrained checkpoints.
Performance
Providers
Page 1 of 1
Alternatives
| 57.09 | - | - | - | - | - | ||
| 55.54 | - | - | - | 2026-07-31 | |||
| 55.22 | 1K | - | - | 2026-03-31 | |||
| 53.81 | - | - | - | - | |||
| 52.55 | - | - | - | - | - | ||
| 51.85 | - | - | - | 2026-02-12 | |||
| 51.84 | 1K | - | - | 2025-10-15 | |||
| 50.96 | 131K | - | - |
ModelRegistry · Catalog updated 2026-11-16
Popular Model Organizations
The organizations behind the models, ordered by model count.
















