The model generates and edits clips of 3 to 15 seconds, in one-second increments, up from a 10-second limit in Omni 1.1. The price is down to $0.07 per second for 720p video. Google says users can draft quickly at 360p or 720p, fix a detail with a follow-up edit instead of regenerating the whole clip, and then upscale the best version to 1080p or 4K.
The model can be guided with a combination of character, product or style images, short motion clips of up to three seconds and audio cues. Google says it keeps faces, garments, packaging typography and brand colors consistent across camera angles, with less drift and morphing between frames, making it suited to virtual try-ons, catalog spins and video ads.