One drawing in · Still a drawing after · Nothing to install
Your drawing moves and stays a drawing — hatching, paper grain and all. Upload a page, get a clip.
A real run on this page's model, from one photographed sketchbook page. Watch the cloak and the clouds rather than the figure — they move while the hatching stays hatching, which is the whole test. Every output is AI-generated, so your result varies with your drawing; treat this as what the tool does rather than a result you will get back identically.
Drag to compare before and after
It animates a still drawing and keeps the drawing intact — the hard part is not the movement, it is holding the medium while things move.
Most tools that claim this quietly do something else: they read the drawing as a description and render a photoreal scene from it. That is a useful thing to have, but it is not what an illustrator asking for this wants. The whole value of the drawing is the drawing — the weight of the line, the direction of the hatching, the places left unfinished.
A clip that replaces all of it with smooth rendered surfaces has thrown away the work and kept only the composition.
So this pass is written the other way round. The clip runs on Seedance 2.5 from your still, with the motion described in the prompt and the medium pinned explicitly: visible strokes, paper grain, no photoreal rendering, camera and page held still. What moves is what would move — a cloak in the wind, hatched clouds crossing the sky, grass at a cliff edge, a head turning a few degrees.
Good sources are drawings with some depth to them: a foreground subject, a middle ground and a horizon give the motion somewhere to happen. A flat character study on white paper has very little for anything to move against.
Creative engine
Upload the drawing, describe what should move, and keep the medium named in the prompt.
Hatching, cross-hatching, smudged shading and paper grain carry through every frame rather than resolving into rendered surfaces.
No rig, no keyframes, no animation software — the drawing you already have is the whole input.
Either side of this pass: turn a photo into a drawing first with the AI sketch generator, then stretch the finished clip further in the AI video extender.
From a photographed page to a clip that still looks hand-made.
Even light, whole drawing in frame, no shadow from your hand and no glare on the paper. Scans work too, and a drawing with foreground and distance gives the motion room.
Name the two or three things that should animate — cloak, clouds, grass — and say the page and camera stay still. Asking for everything at once is what produces drift.
Repeat that it must stay pencil, ink or crayon with paper grain visible. That sentence is what stops the clip resolving into a photoreal scene.
Illustrators, people pitching a shot, and anyone whose sketchbook deserves more than a photo on a feed.
A moving panel answers the question a static one cannot: does this shot hold for three seconds. Faster than an animatic and much faster than a previz build.
A drawing that moves stops the scroll where the same drawing as a still does not, and it stays recognisably your line rather than becoming a render.
The one on the fridge, moving. It is the same pass as everything else here, pointed at the most rewarding possible source — and the AI sketch generator can make a matching drawing from a photo to go with it.
Common questions about animating a still drawing with Kavel.
Not if the prompt says otherwise, and that is the single most important line to include. Name the medium — pencil hatching, ink, crayon, paper grain visible, no photoreal rendering — and it holds. Leave it out and the model will happily resolve your drawing into a rendered scene, because that is the more common request.
One with some depth: a subject in front, something behind it, a horizon or a background plane. That gives the motion somewhere to happen. A single figure on blank paper can only move itself, so the clip has less to do and reads flatter.
Yes. Watercolour, marker, biro and full-colour illustration all animate the same way, and the medium note in the prompt changes accordingly — say watercolour bleed and paper texture rather than graphite hatching, and the same logic applies.
A few seconds per run, which is the natural length for this kind of motion anyway. If you need longer, generate the beat you want and then run it through the video extender rather than asking for one long take up front.
You can ask for a slow push or a gentle drift, but the strongest results on drawings come from holding the camera still and letting the contents move. A moving camera forces the model to invent parts of the page that were never drawn.
Some line breathing is inherent — every frame is generated rather than tweened from the last one. Keeping the requested motion small and specific reduces it a lot, and a source with strong, confident linework survives much better than a faint one.
None. There is no rig, no timeline and no keyframing — a photograph of the page is the entire input, and the clip comes back ready to download.
The estimate is on screen before you generate and moves with the length and resolution you choose. Failed runs return to your balance unless they broke the content policy, so an attempt that came back photoreal has not cost you the credits.
One credit pool covers Nano Banana images and Seedance and Veo video. The cost shows before every run, and failed jobs are not charged. Use a subscription for ongoing work, or a one-time pack when you just need to top up.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
One-time top-ups — buy extra credits any time you run low.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
Charges appear as “KAVEL AI” on your card statement. You can cancel any time from Settings → Billing; cancellation takes effect at the end of the period you already paid for.
Powered by
Use one balance across every supported image and video model — eleven of them today, listed below — and check the credit cost before each request.
Video models
Image models
Model credit guide
You're only charged for successful generations. The exact estimate in the generator varies by model, length, resolution, audio, and number of images.
| Type | Model | Credit cost |
|---|---|---|
| Video | Seedance 2.5 | The long-take tier, priced per second: 5s 480p ≈ 211 credits, 5s 720p ≈ 473. A full 30s take runs ≈ 1,260 at 480p and ≈ 2,835 at 720p. Supplying a reference clip lowers the per-second rate. |
| Video | Seedance 2.0 | 5s 720p image-to-video ≈ 188 credits; text-to-video ≈ 308 credits. Scales with resolution and length. |
| Video | Seedance 2 Fast | Faster and lower cost. 5s 720p text-to-video ≈ 248 credits. |
| Video | Seedance 2 Mini | The cheapest Seedance tier. 5s 720p ≈ 154 credits, 5s 480p ≈ 72 credits. |
| Video | Seedance 1.5 Pro | Audio doubles the rate. 5s 720p ≈ 27 credits silent, ≈ 53 with audio; 1080p ≈ 57 and ≈ 113. |
| Video | Veo 3.1 | Billed per video, not per second (Lite tier). About 45 credits at 720p and 53 at 1080p. |
| Video | Kling 3.0 | Audio raises the rate. 5s 720p ≈ 105 credits silent, ≈ 150 with audio; 1080p ≈ 135 and ≈ 203. |
| Video | MiniMax H3 | Fixed 2K, no resolution ladder. Priced per second — a 5s clip ≈ 158 credits. |
| Image | Nano Banana 2 | Generate or edit from text and images. About 8 credits per 1K image, 12 at 2K, 18 at 4K. |
| Image | Nano Banana Pro | Consistent run times across generations. About 12 credits per 1K or 2K image, 21 at 4K. |
| Image | Nano Banana 2 Lite | Faster, simpler variant. A flat 8 credits per image at every resolution. Start here to test. |
| Image | GPT Image 2 | The lowest-cost image model here. About 3 credits per 1K image, 6 at 2K, 12 at 4K. |
| Image | Seedream 5.0 Lite | Flat pricing — about 8 credits per image whether you generate or edit. |
One photographed drawing in, a clip out that still looks drawn — hatching, paper grain and your own line intact.
