Google has launched Gemini Omni 1.2 Flash, a video generation model that can create and edit clips up to 15 seconds long. It is generally available through the Gemini API and Gemini for business. Google says it has cut the price by 30% from Gemini Omni 1.1, to $0.07 per second for 720p video.
The new clip limit is five seconds longer than Omni 1.1's. Users can choose durations from three to 15 seconds in one-second increments. They can also make follow-up edits to the same clip, such as changing a background, lighting or aspect ratio, without generating it again from scratch.
The model keeps characters and scenes more consistent
Google says the model follows visual and audio references more closely and keeps characters, objects and scenes more consistent across a clip. Users can combine images with motion clips of up to three seconds and audio cues to guide a generation.
The company also claims improvements to dialogue, lip-sync, multiple speakers and continuous camera movements. Google says users can draft at 360p or 720p, then upscale selected clips to 1080p or 4K. Outputs carry an imperceptible SynthID watermark.
Google is rolling out the model across its apps
Developers can use the model through the Gemini API, while businesses can access it through Google Cloud. Google says Provisioned Throughput is available. The model is also rolling out across its apps: Google Vids offers 15-second clips at no additional cost with a Google account or Google Workspace subscription, and access in the Gemini app is rolling out to Google AI Ultra subscribers. Google also lists Flow and Google Ads among the products where it is available.
