AI
Google Gemini Just Made 15-Second AI Videos (30% Cheaper) 🤯
Google’s Gemini Omni 1.2 Flash update brings three changes into the same release: longer generated videos, conversational editing, and a lower price. The SAZ Labs short covers a model that can now produce clips up to 15 seconds long, compared with the previous 10-second limit. That gives a single generation more room to develop an idea before it ends.
Five extra seconds may sound modest, but it changes how much can fit inside one short clip. A creator has more time to introduce a subject, show an action, and let the final image register. The update does not remove the need to decide what the video should communicate. It gives that decision a slightly wider canvas within the same short-form format.
The editing workflow is another part of the announcement. Instead of starting over whenever a clip needs a change, users can edit through a conversation. Multi-turn editing makes revision part of the interaction with the model. The useful distinction is between asking for an entirely new video and continuing to work on the video already being developed.
Google also says faces, products, and brand colors remain more consistent. That is a company claim about the model’s output, rather than a result independently tested in this article. For anyone making a clip around a recognizable person, product, or visual identity, consistency is an important part of deciding whether the result communicates the intended idea.
The price is down 30 percent, to seven cents per second for 720p video. Both the resolution and the per-second basis matter when reading that figure; the brief does not describe pricing for other settings. The model is available through the Gemini API and Google Cloud. Together, the longer duration, continued editing, and reduced price describe an update focused on making short AI video easier to revise and use. Watch the original short below for the compact version of the announcement.
Book a call