Google introduced Gemini Omni 1.2 Flash on November 5, 2026, for creating and editing videos up to 15 seconds long. Finished clips can be refined with conversational prompts without starting over. The model is generally available through the Gemini API and Gemini for business.
Longer clips and conversational editing
Video duration can be set from three to 15 seconds in one-second increments. The previous Omni 1.1 supported clips of up to 10 seconds.
Conversational editing works across a full 15-second video. Users can issue follow-up prompts to replace a product or background, change the lighting or weather, and adjust the aspect ratio.
Google recommends drafting at 360p or 720p, correcting details with further prompts, then upscaling selected results to 1080p or 4K.
Reference material and scene consistency
Video generation can be guided by combining images of characters, products or styles with audio cues and motion clips up to three seconds long. Google says the model better preserves the appearance of characters, objects and scenes from the first frame to the last, including in shots with camera movement.
The company also says consistency has improved when filling in frames between keyframes and extending clips. In a virtual try-on demonstration, Google used a photo of a model and front and back images of three garments to create a 15-second sequence with 360-degree turns.
Dialogue and camera movement
Google promises smoother dialogue, clearer lip-sync and the ability to specify distinct voices for multiple speakers. The company says continuous camera movements, including flythroughs, zoom-outs and orbits, can also be controlled more precisely, with fewer unwanted cuts and steadier scale.
Google says it is continuing to improve fast sports action, rapid script changes and complex physics.
Where the model is available
Developers can prototype with the Gemini API in Google AI Studio. Businesses can try the model in the Google Cloud console and build solutions with the API. Provisioned Throughput is also available, and outputs carry an imperceptible SynthID watermark.
Video at 720p costs $0.07 per second, which Google says is 30% less than Gemini Omni 1.1. Google Vids offers 15-second clips at no additional charge with a Google account or Google Workspace subscription.
In Flow, users can direct scenes and make conversational edits. Creation of 15-second videos in the Gemini app is rolling out to Google AI Ultra subscribers.
