How to make Jean Phil dancing videos
Use the room-dance template and check the movement before sharing.
Preview the dance
Open Templates and play Solo room dance. The preview is an external movement reference; the separate generated examples show actual site outputs. Choose the template to load it in the creator.
Keep the request focused
Use one authorized character image or start without a photo. Request one short dance in one room with a steady full-body camera. Keep feet in frame and avoid asking for many scene changes in a short clip.
Review the full result
Choose 4–15 seconds and 576P or 768P; the displayed credit cost changes with those settings. Watch the full output for foot sliding, distorted hands and changing faces. Add only music you have permission to use; buying credits does not grant rights to the reference soundtrack.
Match duration to one movement phrase
A 5-second request should focus on one readable phrase: a short side step and a pause, for example. Asking for a walk-in, a spin, a jump and a bow adds several competing events. The model may reinterpret the timing even when a video reference is attached. Preview the reference for the action you want; do not choose a template only because its thumbnail looks appealing.
Inspect movement and appearance separately
Watch the first frame, the middle and the final pose, then replay the complete clip. Check whether the feet remain attached to the floor, hands keep a plausible shape, and the body stays inside the frame. Separately compare the input face, hair and clothing. A convincing dance with a different person is not a successful identity-preserving result.
Choose quality for your test budget
For 5 seconds, 576P costs 45 credits and 768P costs 75; for 10 seconds they cost 90 and 150. More pixels are not evidence of better identity retention. Our two published dance tests used different resolutions, but they were not a controlled one-variable comparison and both showed appearance changes. Start from the published limitations rather than assuming a higher price guarantees a matching face.
One template, two real outputs
Both used this AI-created fictional adult input. These tests demonstrate delivery, not accurate identity replacement or a resolution benchmark.
| Stage | Time | Elapsed |
|---|---|---|
| Accepted → rendering | 08:06:50 → 08:13:52 | 7m 02s |
| Rendering → complete | 08:13:52 → 08:18:01 | 4m 09s |
| Total | 08:06:50 → 08:18:01 | 11m 11s |