Gemini Omni 1.2 Flash

Gemini Omni 1.2 Flash is a video generation model that turns prompts and references into clips of up to 15 seconds with synchronized audio. Key capabilities: • Generate clips up to 15 seconds long with native synchronized audio • Edit videos turn by turn through follow-up prompts in conversation • Use up to six reference images, audio, or video to keep characters and products consistent • Extend scenes toward 30 seconds for longer storytelling • Export in 1080p or 4K resolution • Change a scene without restarting the whole generation • Run at a lower cost than the previous 1.1 version. Reference-driven control keeps characters and products consistent from shot to shot. Instead of regenerating an entire clip when something is off, you can request a change such as a different lighting setup or camera angle, and the scene updates while the rest stays intact. Multimodal input lets you combine image, audio, and video references to guide product shots and brand visuals with more precision. Built for marketers, product teams, filmmakers, and social media creators who need fast, cinematic clips without complex editing software. Use it for product showcases, ad concepts, short-form social content, and storyboard previews, then download in 1080p or 4K for publishing.