# VideoMaMa++: Temporally Consistent Video Matting via Preserve-and-Refine Token

> Source: <https://aiflash.com/news/136729/>
> Published: 2026-10-21 13:00:01+00:00

Diffusion-based video matting has recently emerged as a leading approach, leveraging video diffusion priors from internet-scale pretraining to achieve strong zero-shot generalization from limited synthetic data. However, the heavy compute of these diffusion backbones limits inference to a short chun