Video that keeps the face
Most AI generators hold a face together in a single photo. The real test is motion — this is where the character's face stays locked from the first frame to the last.
Face-lock through an entire clip
Drag the slider to compare the start, middle and end of one clip — the same character, holding through every frame.
Face-lock through an entire clip — StartStart
Face-lock through an entire clip — MiddleMiddle
Face-lock through an entire clip — EndEnd
Text to video
Generate a video clip directly from a written prompt, across 23 supported models.
Image to video
Animate an existing photo into a moving clip, across 25 supported models.
Reference to video
Guide the motion and framing with a reference image or clip, across 7 supported models.
| Format | MP4 (WebM also supported) |
|---|---|
| Resolution | 360p – 4K, model-dependent |
| Duration | 1–30 seconds, model-dependent |
| Aspect ratio | 16:9, 9:16, 1:1 and more, model-dependent |
01
Pick a model
Choose a video model based on resolution, duration and style.
02
Write a prompt or upload a photo
Describe the scene, or animate an existing image.
03
Generate and download
The character's face holds through the whole clip, ready to post.
