Today, Google announced the rollout of a new video generation feature in the Gemini app. Powered by its Veo 3 video generation model, this new tool allows users to transform a static image into an imaginative 8-second video. To start creating, locate the Gemini prompt bar > select Video > and upload a photo. Then describe the motion, sound, dialogue, and anything else you envision, and watch Gemini produce an 8-second 720-pixel video with audio and a visible AI-generated watermark.

Starting today, this photo-to-video feature arrives in the Gemini web app, and smartphone users will see the capability later this week. For now, this feature is only available in select countries and to Google AI Pro and Ultra subscribers. And if you’re a user of Google’s AI filmmaking app, Flow, this update is live there too, complete with audio support.