Up to 15-second video generation… Google unveils Gemini Omni 1.2 Flash

Google released Gemini Omni 1.2 Flash, a video-generation model that produces clips of up to 15 seconds in a single pass, up from the 10-second limit of the prior version. The model is now generally available through the Gemini API and Google's enterprise offerings, with developers able to access it in Google AI Studio and creators able to use it in Google Flow for video production. The five-second increase in maximum clip length is the headline change, reducing the number of generations needed to assemble longer sequences. The rollout matters because it pushes Google's multimodal tooling deeper into developer and enterprise workflows, where clip duration and conversational editing are key constraints. What to watch next is adoption across AI Studio and Flow, and whether the longer output window prompts rival model makers to match or exceed the 15-second threshold.