Text-to-video and image-to-video that begin from your stills. One movement per clip; price by duration and resolution, shown before you confirm.
Video is priced by duration and resolution. The exact credit cost is calculated on the job and shown before you confirm; failed jobs refund automatically.
Pick an effect, upload your photo, and the built-in recipe does the rest. Same photo, far more reliable results.
You have been writing prompts by hand. Templates skip all of that — pick a card below, upload your photo, and get the same scene in one click.
NEW 5s
⚡ Generate
Video is the costly end of the studio, so it is aimed rather than scattered. The reliable order is this: a still first — on character, already judged — and then one movement laid over it. Or, for a scene that exists nowhere yet, a text brief that carries the same fixed description as your stills. Pricing follows duration and resolution, is calculated on the job, and is shown in full before you confirm. A failed render refunds itself.
Image-to-video inherits the woman, the dress and the room from the first frame. It is the cheapest consistency there is, and the calmest.
A turn toward the light. A slow walk-in. One movement, camera held. Choreography smears; a single gesture renders clean and loops well.
Decide where the clip will live before rendering it. A teaser and a store piece are different measures, and rendering the maximum of everything is how budgets quietly double.
The week’s strongest frame, given a few seconds of life, to draw a viewer toward the full set.
One clean repeating gesture. It must read at thumbnail size and survive a platform’s compression.
Longer takes, briefed from the same description as the stills they sit beside, so nothing feels cast twice.
“She turns slowly toward the window light and lowers her eyes, hair moving a little, camera still.”
One movement for the subject, none for the camera. The first frame already decided everything else, so the brief only directs motion.
“Fictional 30-year-old woman, long straight black hair in a hime cut, deep indigo wrap dress, crossing a quiet tatami room toward camera, soft grey daylight, unhurried, cinematic.”
The fixed description travels into the clip brief unchanged, so the woman in motion is the woman in the stills.
| Lever | Effect on price | Practice |
|---|---|---|
| Duration | Longer costs more | A teaser rarely needs more than a few seconds |
| Resolution | Higher costs more | Render at the destination’s measure; upscale a winner later |
| Retries | Each render is a job | Correct the brief or the first frame before rendering again |
By duration and resolution, calculated on the job and shown before you confirm. Nothing is charged in the dark, and a failed render refunds automatically.
Image-to-video for series work — the first frame carries the character with it. Text-to-video is for standalone scenes and first experiments.
Yes, by two habits: begin from series stills, and carry the fixed description into text briefs word for word.
Usually too much is asked of it — several movements, a moving camera, a crowded room. One movement, a still camera, an empty space: that is the pattern that holds.
No. Clips rest in your account until you publish or download them. Nothing is posted on your behalf.
Drafts, edits, animation and finishing share one credit balance and one private library. Start with the piece you need today.
Open the generatorYour image is deleted after generation.
Browse filesFailed jobs are refunded automatically. Results land in My Collection.
Sign in with Google to start generating
Opening secure Google sign-in…