The running joke, finally settled
Every household with an animal has a theory about who it would be. Turning the cat to human form ends the argument with a picture instead of another impression at dinner.
One photo in · Same eyes, same colouring · Nothing to install
Turn your pet into a human and find out who has been living in your house. One photo of your cat or dog goes in, and a person comes back wearing its eyes, its colouring and the exact look it gives you at six in the morning.
Made with the model this page uses. Drag the divider — the eyes are the thing to watch: one green, one blue, in the same order on both sides.
Drag to compare before and after
A pet to human generator takes one photo of an animal and rebuilds it as a person, carrying the traits you would recognise across the gap: eye colour and eye pattern, coat colour turned into hair, build, and the expression the animal was actually wearing when the photo was taken. It is not a costume filter and it is not a cartoon — the output is a portrait of a human being who is unmistakably your pet.
Three people upload the same kind of photo for three different reasons. One has spent four years insisting the cat is judging them and wants the receipt. One wants a birthday card for a partner who talks to the dog more than to them. One saw someone do it on X and wants to see their own.
All three want the same thing out of it: a person their friends recognise as that specific animal.
It works in both directions of temperament too. A daft dog produces a person mid-laugh; a cat that has never once been impressed produces someone who looks like they are waiting for you to finish talking. The expression transfers as reliably as the colouring, which is usually the part that makes people send it to the group chat.
You do not need a special photo. A phone snap where the face is lit and mostly facing the camera is enough. What it needs is a clear view of the eyes and the coat, because those are the two things doing the identifying work.
The failure mode worth knowing about is the generic one. Ask for nothing specific and the model will happily hand back an attractive stranger who has no relationship to the animal in your photo — the output looks fine on its own and means nothing to anyone who knows the pet. Everything on this page is arranged to avoid that: the prompt names the traits to carry, and the before/after above is there so you can see what carrying them actually looks like.
Creative engine
Upload a pet photo, pick the shape you want the portrait in, and the credit estimate updates before you press anything. The run happens on this page — the result opens directly under the box.
A white long-haired cat should come back with long white hair; a dog with one ear up should come back with something crooked and alert about the face. Transfer those and everyone who knows the animal sees it instantly.
The example here came from an odd-eyed Angora — one green, one blue — and the person kept both, in the same order. A model that ignored the photo would have given them two eyes the same colour.
From one photo to a person your friends recognise, with the cost shown before you commit.
Face roughly to the camera, in daylight or by a window. The eyes and the coat are what it carries across, so both need to be visible. A sleeping cat gives the model nothing to work with.
The prompt is already filled in and asks for the eye colours in the same order, hair matching the coat, and the same expression. Add anything specific — a torn ear, a white sock, a permanently unimpressed face — and it will come through.
The cost sits beside the button before you press. When the portrait comes back, look at the eyes first: if they match the animal's, the model read your photo. If they do not, run it again with a clearer shot.
People settling an argument, people making a present, and people who just want to know.
Every household with an animal has a theory about who it would be. Turning the cat to human form ends the argument with a picture instead of another impression at dinner.
Birthday cards, mugs, a print for someone who would rather see the dog than another candle. The output is a proper portrait, so it survives being printed.
A share of the people who do this are doing it for an animal that has died, working from the last good photo they have. It is worth saying plainly: it needs one clear picture of the face, and it does not need a recent one. A photo from years ago works exactly as well as one from last week.
What it keeps, what it costs, and what it needs from your photo.
Upload one clear photo of your cat or dog on this page, leave the prompt as it is or add anything specific you want kept, and press generate. The pet to human result appears under the box on this page — you do not go anywhere else and there is nothing to install.
It carries the things that identify the animal: eye colour and pattern, coat colour and length as hair, build, and the expression in the photo. The example on this page came from an odd-eyed cat and kept one green eye and one blue eye in the same order. Markings that are only visible on the body — a white chest, a spotted back — usually do not survive into a head-and-shoulders portrait.
Yes, and cats often read better, because their eye colour tends to be more distinctive and their expression is easier to carry over. Dogs work best when the photo shows the ears and the eyes clearly rather than a wide open mouth.
No, not for the first run. Image tools on this site work for signed-out visitors with a small free balance. An account raises the balance and keeps your history; the tool itself does not change.
The estimate is shown next to the generate button before you press it, and it moves when you change the settings. Nothing is charged for a run you do not start, and a run that fails is refunded.
The eyes, open and in focus, and enough of the coat to read its colour and length. That is the whole requirement, because those two things are what get carried across. A blurred body is fine; a photo taken from directly above, in the dark, or with the animal turned away gives the model nothing to identify.
Anything with a visible face and eyes will produce something. Cats and dogs are the most convincing because the traits people recognise in them — eye colour, coat, expression — map most naturally onto a human portrait.
Yes. Ask for it in the prompt and pick a taller shape before you run it. Portraits are the default because the identifying traits live in the face, and a wider crop spends detail on clothes and background instead.
Yes — take the portrait it produces into a [video model](/video/seedance-2) and it will animate as a person, with sound in the same pass. That is a separate run with its own cost estimate.
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. |
More ways to change a picture of an animal, and the model doing the work.
One photo, and the argument you have been having about that animal for years finally has a picture attached. The cost is on the button before you press it.
Examples made on Kavel with this model.
