Google · Gemini Omni Video
Video that comes with its own soundtrack
Google's multimodal video model in Kubeez. It generates motion AND the audio that goes with it (dialogue, music, ambient), supports named characters for consistency across shots, and accepts a reference video you can transform. Stills below are illustrative; outputs are video files billed in Kubeez credits.
Showcase






























































































































































































































Most video models leave you mixing audio after the fact. Gemini Omni renders both at once—dialogue, score, and ambience inside the same generation pass. Reach for it when the brief is a complete moving moment, not a silent plate.

Lip-sync, room tone, and score arrive in the same render. No second pass, no separate music model, no manual ducking—the audio track is part of what the model produces.

Build a library of recurring characters and reuse them in later generations. Multi-scene continuity that doesn't require remixing the same input clip every time.

Attach a short reference video (up to a 10-second window), describe what changes, and the model rebuilds the shot in your direction—useful for re-styling product footage, b-roll, or rough cuts.
Match resolution, duration, and reference uploads to what the active Gemini Omni card lists. Reference video is capped at a 10-second window; image references max out at 7 (5 if a video is attached).
Hero product moments that arrive with synced VO and ambient bed—skip the audio post and use the render straight in social cuts.

Use named characters for two-shot dialogue, monologue beats, and visual continuity across a multi-shot story without retraining or remixing.

Editorial and music-video frames where the score drives the cut—Gemini Omni renders the picture and the audio that motivates it in a single generation.

Open Media Studio, pick Gemini Omni Video, and generate motion plus sound together.