Wan 2.6 is finally out. It is a new video model with text and image to video up to 1080p generation. It allows for up to 15-second generations. It can handle multi-shot videos. You will be able to import your own audio. This model lets you cast characters from your reference videos to achieve better consistency. It also has native audio to video sync.
Introducing Wan2.6 – A native multimodal model that turns your ideas into breathtaking videos and images!
· Starring: Cast characters from reference videos into new scenes. Support human or human-like figures, enabling complex multi-person and human-object interactions with… pic.twitter.com/lByPJR1Yna
— Wan (@Alibaba_Wan) December 16, 2025
As explained on X, Wan 2.6 gives you control over lens and lighting. You will be able to reference multiple images. The model is already available on Wan’s official website.
