How to Make the Smalltown Boy AI Dance Video With a Friend

A face pressed right up to a wide-angle lens, both hands clutching the head, mouth open in a silent scream, while a friend dances in the background. Then they swap. That's the Smalltown Boy trend, and it's one of the most replayed duo formats on TikTok and Reels right now. Here's what it is, where the song comes from, and how to make your own version from two photos.

Made with apt video ai from two AI sample photos. Photo 1 screams first, photo 2 takes over.

What is the Smalltown Boy trend?

It's a two-person video in four beats:

1. The first scream. One person leans into a wide-angle lens, hands in their hair, mouth wide open. The lens makes the face and hands look huge. The other person dances in the background.
2. The swap. Roles flip: the dancer takes the lens and screams, and the first person dances behind.
3. The backyard dance. The shot cuts wide and the pair dance side by side, with one low-angle shot against the sky.
4. The lunge. Both rush the camera together, hands reaching for the lens, for one last scream.

The joke is the contrast: a dreamy 80s synth track under two people losing it on camera. That's also why it's hard to film yourself. The distortion only works with the right lens held inches from the face, and the screams have to land on the beat.

What song is it?

It's "Smalltown Boy" by Bronski Beat, the British synth-pop band's debut single from 1984, later on the album The Age of Consent. Jimmy Somerville's falsetto over that looping synth line took it into the UK top three, and its story of packing a bag and leaving a town that never understood you made it one of the decade's defining pop records. Four decades on, its opening riff has become the countdown to two people screaming into a camera.

apt video ai is an independent tool and is not affiliated with Bronski Beat, its members, their labels or the original creators of the trend.

How to make the Smalltown Boy AI video

1. Open the Smalltown Boy template.
2. Add two photos. Photo 1 screams first and stands on the left in the wide shots; photo 2 screams second and stands on the right. One person per photo, face sharp and front-facing. You don't need a prompt.
3. Pick a quality and generate. Photos are reviewed first, then the render takes a few minutes. The clip is about 21 seconds. If a photo isn't approved or the render fails, the credits come back automatically.
4. Download or share. The video comes with the song, synced to the picture. On a phone you can send it straight to TikTok, Reels or Shorts.

What the template recreates

The backyard, the fence and the moves come with the template. Faces, hair and outfits follow your photos.

Photo tips

• Sharp and front-facing. The close-ups put each face a hand's width from the lens.
• Nothing over the face. No sunglasses, masks or hands.
• Hair you like. Hair is kept down to length, curl and color, and the hands go right into it during the scream.
• Most dramatic friend in photo 2. That close-up is the longest and ends with the biggest shocked face.
• Use your own photos, or friends who agreed to it. Celebrity photos and film or music video stills won't pass review.

Is it free? What does it cost?

Signing up, uploading and watching the sample are free. Each render costs credits: 15 for 480p, 35 for 720p and 60 for 1080p, with the song included and no watermark. Credit packs don't need a subscription.

Caption ideas

• "Us when the group chat says the trip is cancelled"
• "Me and my bestie finding out it's Monday tomorrow"
• "When the waiter says the kitchen is closed"
• "POV: you both forgot the assignment was due today"

Add #SmalltownBoy and #SmalltownBoyDance and post with the original sound so it's grouped with the rest of the trend.

More templates to try

Want to see what else is trending? Read our AI video trends roundup.