Gemini Omni 1.1 Is Live on Aristotto

Gemini Omni 1.1 Flash is live: Google's video model with real-world physics grounding, keyframe control, and scene extension up to 40 seconds.

Gemini Omni 1.1 Flash is natively multimodal, reading text, image, audio, and video together, and it carries Gemini's world knowledge into every generation. That means scenes follow real-world physics and read as plausible, not just photoreal. Two features make it stand out for iterating on footage you already like: Conversational editing: describe a change in plain language on an existing clip and the model applies it while leaving the rest alone. Short, direct instructions work best, like "make this anime" or "change the lighting to be more dramatic." Scene extension: continue an existing clip from where it left off, reading up to 10 seconds of prior context, in 10-second increments up to a 40-second total, so a story can keep going without an obvious seam. It also supports keyframes: give it an opening and closing image and it generates everything connecting them, which works especially well for camera orbits, zoom transitions, or seamless loops. Prompt tip: iterate fast on a quick 360p draft, then re-run your keeper at 1080p or 4K for final delivery. It's the cheapest way to lock a concept before spending on a full-resolution render. Not sure what to test? A skateboarder ollies off a curb in golden hour light, the board flipping once beneath their feet before they land smoothly and roll away. Camera tracks alongside at a low angle, catching dust kicked up from the pavement. Natural ambient street sounds, wheels rattling against the concrete.