Four independent workflows
Text to video, image to video, video edit, and reference to video use different model inputs and validation rules; select the workflow before adding media.

Video Models
Choose one of four separate Wan 2.7 workflows: text to video, image to video, video edit, or reference to video.
Text to video, image to video, video edit, and reference to video use different model inputs and validation rules; select the workflow before adding media.
Image to video supports a first frame, first plus last frames, or one continuation clip. These source modes cannot be mixed, and a last frame requires a first frame.
Video edit requires a source video. Reference to video requires at least one image or video reference, with no more than five references in total.