Video to video AI

Video to video AI starts from footage you already have. Upload a short clip, describe the new look — more photoreal, a music-video grade, or a style change — and the model redraws it. Motion and camera work stay yours. Unlock with any purchase from $5; no subscription. Technical failures refund credits; clips blocked by safety checks may not.

Describe the new look, not the motion — movement and camera come from your video. Name the style or setting you want changed.

Sign in to upload a video. Everything you have typed is kept.

Upload a video to see the price

New accounts start with 20 free credits. Pricing

How video to video works here

  1. Upload your clip

    MP4 or MOV, up to 15 seconds and 32 MB. The model cannot take sound on clips above 24 fps — which is most phone footage, at 30 fps — so we remove the sound for you at upload. A 4K or very high-bitrate file may also go over the size limit; export it at 1080p.

  2. Describe the new look

    Say what should change: “maximum photorealism, keep the same motion”, “Hollywood music-video lighting”, or “the same street at night in heavy rain”. Keep it to one or two sentences — long pasted essays hit the 900-character limit. You do not need to describe the motion.

  3. Pay by the second of footage

    The price is set by the length of your clip, rounded up to the next second, and shown before you press generate. A 10-second clip is 10 credits.

Specs at a glance

ModelDurationResolutionNative audioReference imagesFrom
Lucy Restyle1–15 seconds720pNoNot used5 credits for 5 seconds

What video to video AI can change

The model redraws every frame of your clip in the look you describe. Paying users often ask for higher photorealism or a cinematic music-video grade while keeping the original motion. Style prompts also work: anime, cartoon, claymation, watercolor, pixel art or a comic-book panel. Setting prompts work too: the same street at night, in snow, in heavy rain, or under neon light.

What stays is the part that is hard to fake: the motion, the camera move, the framing and the timing all come from your footage. That is the difference between video to video and generating from a prompt alone, where the model has to invent the movement.

It restyles the whole frame. It is not a tool for editing one object, removing a background or replacing a face. For a ready-made anime look, use the anime generator; for bold 2D cartoon, use the cartoon page.

Turn a clip into anime

Video to video or image to video?

Use video to video when you already have the performance: a dance, a walk cycle, a product spin, a drone pass. The clip supplies the motion, so the result follows it closely and you only pay for the restyle.

Use image to video when you have a still and want it to move. There the model invents the motion from your description, which gives you more freedom and less control.

A common workflow is to shoot a rough take on a phone, trim it to the few seconds that matter, and restyle that. It costs less than generating the same shot from text several times until the movement is right.

Animate a still with image to video

What a video to video render costs

Lucy Restyle is billed at 1 credit per second of the clip you upload, rounded up. A 5-second clip is 5 credits and the 15-second maximum is 15 credits, which makes this the cheapest mode on the site.

The price is fixed by the length of your upload, not by the prompt or the number of words in it, and it is shown next to the button before anything is charged. If a render fails for a technical reason, the credits go back to your balance automatically.

See credit packs and plans

Credits for video to video

Lucy Restyle on this page does not spend Seedance welcome credits. Those apply to Seedance Mini drafts on the free trial page.

Video to video unlocks with any purchase from $5, then bills 1 credit per second of your upload. If you only need a Seedance text or image draft first, start on the free trial page and come back here to restyle.

Seedance 2.0 free trial details

Cartoon and other style looks

Anime is one restyle. The same video-to-video engine also does bright 2D cartoon, claymation, watercolor, pixel art or a comic-book panel — anything you can describe in a sentence while keeping the original motion.

Use the cartoon page when you want that look prefilled. Stay on this page when you want a custom prompt, a photoreal polish, or a setting change such as night rain or neon light.

Style restyles redraw the whole frame. They do not swap one character for another or cut out a background.

Turn a clip into a cartoon

Getting a clean restyle

Describe the style or setting, not the action. The motion is already in your clip.

One clear change per render works better than five. Prefer a short prompt over a long ChatGPT essay — this model caps at 900 characters.

Start with a clip where the subject is large and well lit. Small, fast or heavily compressed subjects are harder to follow.

Trim to the part you need. You pay per second, and shorter clips come back faster.

Video to video questions

Other ways in