Gemini will turn a still photo into a short moving clip, and it does the work from one prompt. Open the tool menu in the prompt box, choose Videos, and upload the picture. One video and up to five photos can go in with the prompt, and you can start from a template instead of a blank prompt. 12
Then say what should happen in the scene and what it should sound like. The model behind this is Gemini Omni, which makes clips of about ten seconds with native audio. The sound is part of the prompt itself, and a generation takes a few minutes. If you upload a picture, the video takes that picture's shape, and a prompt with no picture comes back in landscape. 12
Afterwards you keep working in the same conversation: ask for a new camera angle, or to swap, remove or restyle something in the shot. When the clip is right, Share saves it to your device, exports it, or posts it to YouTube. 12
The plan is what decides it. On a personal account, video generation needs a Google AI plan — AI Plus, AI Pro or AI Ultra. A work or school account needs a qualifying Workspace licence, and the feature is for adults 18 and over. 13
Media generation such as video also draws on your usage limit faster than ordinary chat, and every video Gemini makes carries Google's invisible SynthID watermark. 23
It earns its place when you already have the picture — a product shot, a drawing, a view you want to see move. Ten seconds is enough. For a longer piece, ask Gemini what happens next and extend the scene, one clip at a time. 12