VellmHow to make the Smalltown Boy cat AI video with your cat
By Vellm ·
The Smalltown Boy cat video is a duet. One cat stands right up against the camera, mouth wide open as if it is belting the high note, while a second cat dances on its hind legs in the background. Then they swap places. At the end, both stand side by side and dance together. It is named after the song it is posted with, Bronski Beat's "Smalltown Boy".
To make it with your cat, open the Smalltown Boy template, choose Pets, and upload one or two cat photos. You get a 21-second vertical video (9:16) for 100 credits. There is also a People mode for two friends.
What happens, second by second
Here is how it plays out in Vellm's example, with a long-haired tabby and a black-and-white cat:
| Time | What you see |
|---|---|
| 0–3.75 s | The tabby screams into the camera in close-up. The black-and-white cat dances behind it. |
| 3.75–5.5 s | The tabby pushes its paws right at the lens. |
| 5.5–6.75 s | The tabby settles back, paws together, while the other cat keeps dancing. |
| 6.75–14 s | The swap: the black-and-white cat takes the close-up, paws on its head, screaming. The tabby dances behind. |
| 14–19.5 s | Both cats stand side by side on their hind legs and dance together, with a closer angle for a moment. |
| 19.5–21 s | They reach toward the camera together to finish. |
The swap at 6.75 seconds is the turn of the clip. Each cat gets roughly seven seconds of close-up, so both need a clear face in their photo.
One cat or two
With Pets, Cat A screams first and Cat B dances behind, then they swap. Only Cat A is required. If you leave Cat B empty, AI adds a dancer cat, so you still get the duet rather than a solo.
If you have two cats, give Cat A to the one with the more expressive face. It owns the first close-up.
For the photos:
- One cat per photo, face clear and well lit.
- The original photo, not a screenshot, and no filters that hide markings. Vellm's example used a tabby sitting on a rug and a black-and-white cat lying on a sofa.
- Your cat does not need to be standing. The template makes the dance.
People mode
Choose People to make it with two friends instead. Person A screams first, Person B dances behind. People mode needs two photos, one adult in each, who have agreed to appear. No celebrities and no one under 18.
Scene, sound and price
Before you generate, the template lets you pick a scene; the routine stays the same in each. The video takes about 10–15 minutes, and a one-time purchase needs no account. When it is ready, watch the swap at 6.75 seconds: both characters should still look like themselves as they trade places.
Post the silent download and add the Smalltown Boy sound inside TikTok or Reels. With Pets you also get a download with the original clip's sound, for personal viewing only.
It is a 21-second video, so it costs 100 credits whether you use one cat or two. The smallest pack that covers it is 100 credits for $9.99, with no subscription; pricing has the larger packs.
For more cat videos, see the AI cat dance page.