Create your first AI video

This guide walks you through making a short clip in AI Video. It only takes a few minutes, and you can try it with your free sign-up credits.

Step by step

1. Open AI Video from the menu on the left.
2. Choose a tab. Generate Video works from a text prompt, with optional reference images, videos or audio. Frame to Video animates an image you upload as the first frame of the video.
3. Describe the scene. Say who or what is in it, what happens, and how the camera moves — for example: “A red fox walks through a snowy forest at sunrise, slow tracking shot.”
4. Pick a model. Wan 3.0 is selected by default: it makes 2 to 30 seconds of video up to 1080p, with sound. See Compare video models for the others.
5. Set the options such as duration, resolution, aspect ratio and Include audio in video. The credit cost updates as you change them.
6. Press Generate. Most clips take a few minutes. You can leave the page while you wait; the result is saved to My Creations.

Tips for a good first result

• Start short and small (for example 4–5 seconds at 480p), then make the version you like in higher resolution.
• Give each scene enough time. Many scene changes in a few seconds rarely work well; for a story with several scenes, use a longer clip or Super Director.
• With Frame to Video, use a sharp image with the same aspect ratio you want for the video. Turn on Add end frame to set the last picture too.
• Not sure what to write? Try Super Director: describe your idea in a chat and it plans the scenes and makes the video for you. Planning is free with the default writing models; paid writing models and media generation use credits separately.