Wan 3.0 Reference-to-Video

Combine reference images, videos, and audio into one coherent video

Wan 3.0 Prime Reference-to-Video

Reference images (up to 10), videos (up to 5, ≤15s total, ≥16fps), and audio (up to 5).
Follows character identity, visual style, movement, and sound cues from references.
Reference media positionally, e.g. “the subject in Image 1 walks past Video 1”.
Resolution: 480p, 720p, 1080p. Leave duration blank for smart duration.

Prompt

Reference Images (up to 10)

Reference Videos (up to 5, ≤15s total, ≥16fps)

Cloned Voices (adds as reference Audio 1, Audio 2, …)

Reference Audio (up to 5)

Resolution

Aspect Ratio

Duration (seconds, blank = smart)

Seed (optional)

Generate audio

Enable prompt expansion

Enable thinking (advanced reasoning)

Enable safety checker

Download