How to make a K/DA Firestarter AI video

Last updated: October 24, 2026

Use one photo with the supplied K/DA Firestarter reference to make a 5-, 12- or 14-second dance remix. Landscape 16:9 is the default.

1. Prepare a photo

Use a clear JPG, PNG or WebP photo of one consenting adult facing the camera, with the head and shoulders visible and reasonably even lighting. Avoid group shots, sunglasses, masks and heavy filters because they remove details the model may need.

2. Choose the reference segment

Choose 5, 12 or 14 seconds at one credit per second. The opening runs from a face close-up pulling back to the upper body, followed by the arm sweep and turn. The 14-second option covers the full source, including the inverted camera ending; shorter options stop before it.

3. Choose a format

Landscape 16:9 is the default for the supplied reference. Select portrait 9:16 when you need a taller social post. The format changes the frame, while the reference timing stays tied to the selected duration.

4. Generate, check and share

Sign in, confirm that you have permission to use the photo, and submit the render. Queue time depends on service load and length. When the MP4 is ready, download it and keep the AI label when sharing.

Your K/DA Firestarter moment is one photo away

Create my video