The before and after
Real photo, then the paper version, captioned with whatever the cartoon is clearly thinking. It is the format the trend is built on.
Upload one selfie and get the deadpan paper-cutout you, your hair and jacket included, ready for the group chat.
Drag the divider. On one side, an ordinary selfie on a city street. On the other, the same woman as a paper-cutout kid at a mountain bus stop: same curls, same glasses, same green parka, a stare that says the bus is late again.


Drag to compare before and after
The South Park AI filter is the TikTok trend of turning a real photo into a flat, construction-paper cartoon kid in the style of the long-running animated show, usually posted as the plain photo first and the cartoon second.
Upload a photo, tap a scene, and get an original paper-cutout character that keeps your hair, glasses, skin tone and clothing colours. A new account starts with free credits, and weekly check-ins add more. The look is a big round head, white oval eyes with dot pupils, a one-line mouth and mitten hands; the scene, the pose and the expression are words in the prompt you can change.
On TikTok the trend runs on CapCut templates and "how to do the South Park AI filter" tutorials; in the week to 5 October two of those tutorials had passed 1.1 million plays between them.
Creative engine
Upload a photo, tap a scene or describe your own, and get yourself back in paper.

Red curls, tortoiseshell glasses, an olive parka on a grey autumn street. That one photo is the whole input for three of the scenes below.
The curls become a mop of torn orange paper, the glasses stay, the parka keeps its pockets, and she stands at a snowy bus stop with a perfectly flat stare.
Want more than one cartoon of yourself from the same photo? The Multiaverso AI filter makes the cartoon, the kid and the older you. Want a love story instead of a bus stop? The Disney boyfriend filter draws one for you.
The sidewalk selfie from the top of the page, sent into two more scenes. The curls and the glasses carry over every time; only the setting and the drawing style change.

A wooden school desk, a blank green chalkboard, a clock on the wall and one raised eyebrow. The scene for anyone who was always about to say something.
Send me to class
An even cruder drawing: a round head split by the mouth so the top flips open like a lid when you laugh, two bead eyes, stick arms. The one people put in the comments to start an argument.
Flip my lidUpload a photo with a friend in it and both of you get drawn. Click through: the twists, the gold stud, the bob with bangs and the striped top survive into every scene.



The filter works on pets as well. A corgi in a red bandana on the lawn goes in; a paper-cutout corgi with the same big ears and the same bandana comes out, wagging in the snow.


Drag to compare before and after
No template to hunt down, no app to install, and the picture is yours to keep and post.
Front camera or a friend's shot, decent light, nothing covering your face. Glasses, hats and bright jackets are good: they are what makes the cartoon read as you. Group photos and pet photos work the same way.
Pick one of the six above the prompt box: the bus stop, the classroom, the flap-head, the snowman, the couch or the dog. Each loads the full prompt behind the picture on this page. Swap the snowy street for your office, or the deadpan stare for a scream, and the result follows.
The post that works is the real photo first and the cartoon second. If a specific detail like freckles or a beard did not carry over, add it to the prompt and run it again.
The cartoon is the setup; the caption is why it gets shared. The versions that travel furthest:
Real photo, then the paper version, captioned with whatever the cartoon is clearly thinking. It is the format the trend is built on.
Everyone sends a selfie, you run the same scene on each, and the group chat votes on who came out most like themselves.
The bus stop scene crops cleanly to a square: a flat cartoon you is easier to spot in a list of avatars than a photo.
Change the scene to your team's desks and run it on the work photo. Nobody in the team chat is safe after that.
Common questions about the South Park AI filter and making yours on Kavel.
Upload the photo on this page, tap one of the six scenes and generate. The editor redraws you as a flat construction-paper cartoon kid, keeping your hair, skin tone, glasses and clothing colours, and you download the picture when it is done.
Yes, this page does it from your photo. Character makers like SP-Studio work the other way: you build the character by picking hair, eyes and clothes from menus, which is handy if you want full control and no photo at all.
Yes. The filter keeps what is in your photo, so long hair, earrings, makeup and your outfit carry into the cartoon. The bus stop example on this page is a woman with red curls and glasses.
On TikTok the South Park AI filter is mostly a CapCut template: you open the template from the video and add your photo. Here you skip the template and the app: upload the photo in your browser, pick a scene and download the image to post anywhere.
A new Kavel account starts with free credits, enough to try it on your own photo, and weekly check-ins add more. After that, credit packs and plans cover more runs.
No. Kavel is not affiliated with South Park or its makers, and nothing here draws a character from the show: every result is an original character of you in a construction-paper cutout style.
Credits are shared across everything Kavel runs. You see the cost before each run, and a failed job is never charged.
Pay safely and securely with
Payments are processed by Stripe over an encrypted checkout. Your card details are handled by Stripe and are never stored by Kavel.
Upload one photo, tap a scene and post the paper version of you before anyone else in the group chat does. A new account starts with free credits, and weekly check-ins add more.