Multimodal Reference Suite
Combine reference images, motion clips, audio tracks, and storyboard documents in one request.
Direct AI video generation using reference images, video clips, audio tracks, and storyboard documents with Wan 3.0. Maintain character consistency and motion style.
Optional · up to 10 JPG, PNG, BMP, or WEBP images · 10MB each · Min 240px
Optional · up to 5 videos · 100MB each · 15 seconds total
Optional · up to 5 audio files · 15MB each · 15 seconds total
Combine reference images, motion clips, audio tracks, and storyboard documents in one request.
Borrow specific camera motions, pacing, or character movements from an existing reference video.
Ensure uniform aesthetic tone, color grading, and lighting across diverse generated scenes.
Upload images for identity, video for movement, audio for rhythm or sound and one document for longer creative direction.
References should reinforce the same subject, visual direction and intended movement.
Tell Wan 3.0 which identity, motion, style, sound or narrative element to follow from each reference.
Reference mode accepts up to 10 images, 5 short video clips, 5 audio files, 1 document, and 1 public URL reference.
Turn your idea into a cinematic video with synchronized native audio in minutes.