Vidu Q4 Reference to Video
Uses up to 12 image references and three voice clips for consistent subjects and voices in 3–16 second videos, up to 4K.
Edit any video with a natural-language instruction using Bernini-R, changing objects, weather, background, or camera angle while keeping the rest of the scene intact.
Scores appear here when an evaluation matches this exact model.
fal-ai/bernini-r/edit-videoUpdated Oct 20, 2026
Uses up to 12 image references and three voice clips for consistent subjects and voices in 3–16 second videos, up to 4K.
Vidu Q4 turns a first-frame image into 3 to 16 second videos with native audio at up to 4K. Best for animating stills, product shots and character scenes with dialogue.
Kandinsky 6.0 VSR Lite is a fast, lightweight video super-resolution model for quick, cost-efficient upscaling.
Kandinsky 6.0 Pro is Kandinsky Lab's flagship text-to-video model, generating high-resolution clips with cinematic motion and precise prompt adherence.