An updated version of Google DeepMind's Gemini Omni 1.2 Flash video model has generated clips up to 15 seconds since Nov. 5, up from 10, while the price of 720p video dropped 30% to a rate of $0.07 a second, the company said.
The update adds a conversational editing feature that lets users adjust a single detail in a clip without regenerating the whole video. Google also improved lip-sync with dialogue and smoothed camera movement, and clip length can now be set in one-second increments from 3 to 15 seconds. Every generated video carries an invisible SynthID watermark.
The model is available through the Gemini API, Google AI Studio and Google Cloud Console, and is built into the Flow editor, Google Vids, the Gemini app and Google Ads. Google lists advertising clips, virtual clothing try-ons and multi-character dialogue scenes among the use cases.


