Gemini now allows users to create videos from a single image. On Thursday, Google unveiled image-to-video capabilities with the Veo 3 video generator in the Gemini app. Users can upload an image to Gemini and transform it into a video with audio using straightforward text prompts. Image-to-video tools such as Luma and Kling have gained popularity for animating memes or bringing still images to life. Now, Google provides this functionality with Gemini.

Veo 3, which was launched at Google I/O in May, powers this feature. Unlike previous versions, Veo 3 offers audio support, quickly attracting attention online. AI-generated videos of news reports and street interviews have gone viral due to their realism and issues concerning misinformation. Google stated that all videos produced by Veo 3 carry a visible watermark and an invisible SynthID watermark. Subscribers to AI Pro or AI Ultra will find the “Videos” option in the tool menu, enabling them to upload an image and provide a description in the prompt box.

In the demo video released with the announcement, an image of a cardboard box evolves into a video showcasing various contents, including an elevator, a model ship on water, a rock concert, and a mouse in a tiny kitchen. Google reported that over 40 million videos have been generated via Veo 3 on the Gemini app and Flow, its AI filmmaking platform.

To use this tool, users need a paid subscription to Google AI Pro, priced at $20 per month, or Google AI Ultra, at $250 per month. Users must be over 18 years old and possess a personal Google account, as it is not compatible with corporate or educational accounts.