Smalltown Boy AI: The Screaming Cat Dance and How to Put Your Own Pets in It

A bright, minimal living room, filmed from down low with a wide-angle lens. A grey cat stands on its hind legs right in front of the camera, clamps both paws on top of its head and screams. Behind it, a spotted cat is dancing like nobody is watching. Then the two of them line up side by side, dance, swap places and charge the camera until their faces fill the whole frame.

That is the Smalltown Boy AI cat dance, and this month it has been impossible to scroll past. This post covers where it came from, how the clip goes, and how to make a Smalltown Boy AI video with your own two pets, or with you and your pet.

What is the Smalltown Boy AI trend?

It is a short AI-made clip of two cats, and it spread in early October. One of the first versions went up on October 2, and a second one posted on October 8 passed a million views in three days. Within a week people were remaking it with their own cats, their dogs, their siblings and themselves.

To be clear, no real cat did this. The cats in the original are AI-generated, which is exactly why everyone feels free to swap in their own. The joke works because of the contrast: one cat is having a full meltdown at the lens while the other is calmly having the best night of its life.

People search for it as "smalltown boy ai" or "smalltown boy cats", and a Smalltown Boy AI clip with your own pets is what most of them are after.

The clip, shot by shot

Our Smalltown Boy Cat Dance template uses the part of the clip everyone copies: 14 seconds, 9:16 vertical, five shots.

The scream (first 7 seconds): the front one holds its head and screams into the wide-angle lens while the other dances behind it.
Side by side: the two line up in a wide shot and dance together.
Swaying: still side by side, leaning into each other.
The swap: they trade places and keep dancing.
The rush: both charge the camera until their faces fill the frame, one paw reaching out.

Five shots from a Smalltown Boy AI clip: the scream into the lens, dancing side by side, swaying, swapping places, rushing the camera

Here is the scream. The animations in this post are silent; the real clip has sound.

A woman in a mustard sweater holds her head and screams into the lens while an orange tabby cat dances on its hind legs behind her

Picking the two photos

For the cat dance, the template takes exactly two photos, and the order matters:

Photo 1 screams at the camera. It gets the long close-up at the start, so its face carries the clip. Pick a photo where the face is clear and looking roughly at the camera.
Photo 2 dances behind, then joins in for the side-by-side part. A photo that shows the whole body helps.
One subject per photo. In a group photo the template cannot tell who to use.

Any pair works: two cats, a cat and a dog, two dogs. Each pet stays a real animal with its own face, fur and markings, standing on its hind legs like the cats in the original.

You can also put yourself in. Next to each photo there is a choice, Pet or Person. Set photo 1 to Person and you are the one losing it at the camera while your cat dances behind you. Adults only.

Photos next to results: a woman and an orange tabby, a ragdoll cat and a French bulldog

Only use photos of people who are happy to appear. The result looks like you rather than a copy of you: friends will recognize you right away, but it will not match the photo frame for frame.

How to make a Smalltown Boy AI video with ZOOOP

Open the template page and upload photo 1 (the screamer) and photo 2 (the dancer).
Set each one to Pet or Person.
480p is the default; pick 720p if you want it sharper.
Hit generate. It usually takes five or six minutes.

No prompt and no editing. The room, the camera, the five shots and the timing are already in place.

Here are the ragdoll and the French bulldog in the side-by-side part.

A ragdoll cat and a French bulldog dance side by side on their hind legs in the bright living room, then swap places

What surprised us while building the template

In the side-by-side shots, a woman in a mustard sweater comes out the same height as the orange tabby next to her, with short legs and bare feet

People shrink to pet size. In the original, the two dancers are the same height, and the template keeps it that way. So when one photo is a person, the side-by-side shots squeeze them down to the height of the pet next to them, short legs, bare feet and all. Or the cat grows to human size, depending on how you look at it. We thought about fixing it and decided not to. It is funnier like this.

The last shot is supposed to look stretched. In the final rush, noses, paws and hands look huge as they come at the lens. That is the wide-angle lens doing its thing, and the original looks the same.

Pets are less predictable than people. That has been true in every template we have built. If a pet's face or markings come out a little off, run it again before you change the photo.

More ways to play

Make one with every pet in the house and see who screams best. For a dance with just one pet, try Gang Gang Dance or Dodge Step Dance, and for a pet with more drama, Knight Cat turns your cat into a tiny knight in a storm. Everything that turns a photo into a clip lives in the template library, and if you would rather write your own scene, start from AI video generation.

Pick your two loudest pets and make your own Smalltown Boy AI video with the Smalltown Boy Cat Dance template. Paws on head, mouth wide open.