Google’s updated AI video maker creates 15-second clips and cuts its price 30%
Gemini Omni 1.2 Flash lets users edit a clip by asking for changes in plain language instead of starting over.
Google on Thursday released Gemini Omni 1.2 Flash, an updated AI model that makes short videos from written descriptions and pictures, and cut its price by 30%, according to the company’s announcement.
With AI video tools, a user types a description, called a prompt, and can add photos, short motion clips or audio as references. The model then generates brand-new video frames to match. Google said version 1.2 makes clips up to 15 seconds long, up from 10 seconds in version 1.1, with any length from 3 to 15 seconds.
The biggest change for users is editing by conversation. Instead of regenerating the whole clip, a user can ask the model to swap the background or change the lighting and keep the rest. Google says faces, clothing and product labels now stay more consistent from the first frame to the last, camera moves have fewer unwanted jump cuts, and speaking characters’ lips match their words more closely.
For app developers using Google’s Gemini API, the price is 7 cents per second of 720p video, or about $1.05 for a 15-second clip. Users can draft at lower quality, then upscale the best take to 1080p or 4K.
For everyday users, Google says 15-second clips are free in Google Vids, its video-editing app, with a Google account, and are rolling out in the Gemini app to Google AI Ultra subscribers. Every video carries SynthID, an invisible watermark that lets software flag it as AI-made. Google says the model still struggles with fast sports action and complex physics.
Source: Google, Nov. 5, 2026