Google has introduced Gemini Omni 1.2 Flash, an updated AI video generation model that it says costs 30 percent less than its predecessor.
Users can now create videos of 3 to 15 seconds in one-second steps, up from a previous limit of 10 seconds. Generating 720p video costs $0.07 per second through the Gemini API and Google Cloud.
The model lets users combine character, product and style images with motion clips of up to three seconds and audio cues. Google says it keeps details such as facial features, clothing, product packaging, typography and brand colors consistent across camera angles and scenes.
People can also edit an existing video through follow-up instructions instead of generating a new clip. Edits include replacing products or backgrounds, adjusting lighting and weather, and changing aspect ratios.
Google says the model offers smoother dialogue, better lip-syncing, more controlled camera movement and fewer unwanted scene transitions and visual inconsistencies. Clips can first be generated at 360p or 720p and then upscaled to 1080p or 4K.
Gemini Omni 1.2 Flash is rolling out in Google AI Studio, Gemini for business, Flow and Google Vids. Fifteen-second generation is also coming to the Gemini app for Google AI Ultra subscribers. Google says generated videos carry an imperceptible SynthID watermark.
