SMALLTOWN BOY AI VIDEO · TWO PHOTOS
Smalltown Boy AI Video Generator
Upload two photos, one person each, to make a 9:16 Smalltown Boy AI video. Photo 1 screams into the ultra-wide lens first (0–7 s) and dances on the left; Photo 2 takes the second close-up (7–14 s) and dances on the right. Pick 5, 10 or 15 seconds on any Seedance model, or the full 24-second routine on Seedance 2.5 — no prompt needed.
Two photos · 5–24 seconds · 9:16 · Smalltown Boy track included · From 14 credits; the full 24 s from 245 credits (Seedance 2.5, 480p) · Free accounts get a moving watermark; any purchase unlocks watermark-free downloads · Failed generations are refunded
Photo 1 · Screams first
One person, face and hair clearly visible
Photo 2 · Screams second
One person, face and hair clearly visible
Each photo: 300–6000 px wide and high, up to 30 MB. Use a clear, front-facing photo of yourself or a friend who agreed to be in it; waist-up or full-body shows the outfit best. Celebrity photos and film stills are rejected by Seedance 2.5's review.
480p · 9:16 · MP4 · Smalltown Boy track included
This video costs 245 credits. Charged when you start, refunded automatically if it fails.
Smalltown Boy AI video examples
Explore remakes of the trend. The cat versions aren't supported here yet; your video follows the same screams, swap and dance with your two photos.
What is the Smalltown Boy AI video?
A Smalltown Boy AI video puts two people from two photos into the viral Smalltown Boy dance: each takes a turn screaming into an ultra-wide lens with hands on their head, then both dance side by side in a courtyard and lunge at the camera, set to Bronski Beat's 1984 song. On TryH3, Photo 1 screams first (0–7 s) and dances on the left, Photo 2 takes the second close-up (7–14 s) and dances on the right. Each photo first becomes a character sheet, then Seedance recreates the routine from a fixed motion reference — no prompt. Choose 5, 10 or 15 seconds (the opening) on any Seedance model from 14 credits, or the full 24 seconds on Seedance 2.5 for 245 credits (480p), 542 (720p) or 954 (1080p). Every video is a 9:16 MP4 with the song.
Quick facts
| Photos needed | 2 photos total — one person in each photo |
|---|---|
| Photo 1 role | First screaming close-up, 0–7 seconds; left side of the dance |
| Photo 2 role | Second screaming close-up, 7–14 seconds; right side of the dance |
| Character consistency | Each photo first becomes an AI character sheet (front, back, profile and outfit views) so the face, hair and clothes carry into every shot; photo order sets the roles, not gender |
| Motion | A fixed 24-second dance reference sets the camera, cuts and choreography |
| Background | The original courtyard (default), a red fence garden, a living room or your own uploaded photo; the moves and camera stay the same |
| AI model | Seedance 2.5 (5–24 s; the only model that makes the full 24 s) or Seedance 2.0 Mini, Fast and Pro (5–15 s), reference-guided |
| Length and format | 5, 10 or 15 seconds (the opening, with both screaming close-ups from 10 s) on every model; the full 24-second routine on Seedance 2.5; vertical 9:16 MP4 |
| Resolutions | 480p (480×854, default) and 720p (720×1280) on every model; 1080p (1080×1920) on Seedance 2.0 Pro and 2.5 |
| Music | The reference's Smalltown Boy track is laid under the video, so the MP4 has the song |
| Prompt | None — upload two photos and pick a resolution |
| Price | From 14 credits (Seedance 2.0 Mini, 5 s, 480p); the full 24 s on Seedance 2.5 is 245 credits (480p), 542 (720p) or 954 (1080p); character sheets included, price shown before you press Generate |
| $5 Trial pack | 24 credits, one-time: covers a 5 or 10 s 480p Seedance 2.0 Mini video (14 or 19 credits) or a 5 s 720p one (20); the full 24 s starts at 245 credits, and the Creator pack ($49.90, 1,280 credits) covers five |
| Watermark | Moving tryh3.com mark on videos made before your first purchase; none after any credit pack, and earlier videos unlock |
| Failed or rejected | All credits come back if a generation fails or Seedance 2.5's provider review rejects a photo |
| Pets | People only in this version; the two-cat version is not supported yet |
The Smalltown Boy dance, shot by shot
Six moments in the full 24 seconds, with the same cuts as the reference. Shorter videos are the opening: 5 s is Photo 1's scream, 10 s adds the swap, 15 s ends as the dance begins.
- 0:00
Photo 1 screams into the lens
An extreme ultra-wide close-up: both hands on the head, mouth wide open, while Photo 2 dances in the background on the left.
- 0:03
Palms at the camera
Photo 1 shoves both open hands at the lens, then leans in wide-eyed.
- 0:07
The swap
Hard cut: Photo 2 takes the close-up, hands on the head, screaming in shock, then sliding the hands down to the cheeks. Photo 1 dances behind on the right.
- 0:14
Courtyard dance
Wide shot from slightly above: the two face each other and swing and cross their arms in a quick hand-jive — Photo 1 on the left.
- 0:16
Low angle
The camera drops to the ground and looks up as both bounce in a half-crouch, throwing their arms up against the sky.
- 0:18
Frozen, then the lunge
Both stand wide-eyed, mouths open, then rush the lens together with their hands reaching for the camera to end it.
How to make a Smalltown Boy AI video with 2 photos
No filming, prompt or editing app: upload one clear photo of each person, check the price and generate the whole 24-second dance.
- 01
Add two photos
Photo 1 is whoever screams first and stands on the left; Photo 2 takes the second close-up and stands on the right. One person per photo, face and hair visible.
- 02
Pick a model, length and resolution
5, 10 or 15 seconds on any Seedance model, or the full 24 seconds on Seedance 2.5 (the default); 480p is the default resolution. The price appears next to the Generate button. Each photo becomes a character sheet, then the video renders.
- 03
Download and post
You get a 9:16 MP4 with the Smalltown Boy track, ready for TikTok, Instagram Reels and YouTube Shorts.
Photo 1 vs Photo 2: who screams first?
The upload order decides the roles, and your two photos play them in every shot — close-ups, background and the dance.
- Order, not gender. Two guys, two girls or a couple all work; the reference's man and woman don't decide who plays what.
- Mixed them up? Use the Swap button before you generate to flip Photo 1 and Photo 2.
- Put the bigger reaction first. Whoever has the best scream face makes the opening land.
Why each photo becomes a character sheet first
The dance is filmed with an ultra-wide lens inches from each face, then cuts to full-body shots from above and below. One selfie doesn't show your back, your profile or your shoes, so TryH3 first turns each photo into a character sheet — front and back full body, a large portrait and three head angles — and gives the video model both the sheet and your photo. The photo decides the face; the sheet decides the body and outfit.
Your look comes only from your images. Long hair stays long, a hat stays on, glasses stay — and a shaved or bald head stays bald. If your photo stops at the shoulders, the sheet continues your visible clothes into a simple matching outfit.
From the dance reference
- Camera. The ultra-wide close-ups, the low angle and the handheld moves.
- Cuts and timing. The swap at seven seconds and every beat after it.
- Setting. The slatted wooden fence with ivy, the stone courtyard and the daylight.
From your photos
- Faces. Each person keeps their own face, even stretched by the wide lens.
- Hair or head. Hairstyle, colour and length — or a bald head — exactly as in the photo.
- Clothes. Outfit, shoes, hats, glasses and jewellery from each person's sheet.
Price by model, length and resolution
Seedance 2.5 is the only model that makes the full 24 seconds in one run; Seedance 2.0 Mini, Fast and Pro make the 5, 10 and 15 s openings. Prices are in credits and include both character sheets.
| Model | Resolution | 5s | 10s | 15s | 24s | Notes |
|---|---|---|---|---|---|---|
| Seedance 2.5 | 480p | 58 | 107 | 157 | 245 | Default; the only model for the full 24 s |
| 720p | 120 | 231 | 342 | 542 | ||
| 1080p | 206 | 403 | 600 | 954 | ||
| Seedance 2.0 Mini | 480p | 14 | 19 | 25 | — | Cheapest; 5–15 s, 480p or 720p |
| 720p | 20 | 32 | 44 | — | ||
| Seedance 2.0 Fast | 480p | 29 | 49 | 69 | — | Faster; 5–15 s, 480p or 720p |
| 720p | 52 | 96 | 140 | — | ||
| Seedance 2.0 Pro | 480p | 43 | 77 | 111 | — | 5–15 s, up to 1080p |
| 720p | 82 | 156 | 229 | — | ||
| 1080p | 193 | 378 | 563 | — |
At the Creator pack rate ($49.90 for 1,280 credits), the full 24 s at 480p on Seedance 2.5 uses about $9.55 of credits — a credit cost, not a single-video checkout price. Results vary between runs and models; failed generations cost nothing.
Does the Smalltown Boy AI video include the song?
Yes. Seedance 2.5 renders a silent video, then the reference clip's Smalltown Boy track is laid underneath, so the screams and the swap land on the music in the MP4 you download.
The recording belongs to its rights holders. If you post publicly the platform may match it; you can also mute the MP4 and add "Smalltown Boy – Bronski Beat" from TikTok's or Instagram's own sound library.
What is the Smalltown Boy dance trend?
The Smalltown Boy dance started as a real video: on 4 August 2026, Elvira (@elvira.mua.music) and dancer Siarhei Dratvitski (@mr_unvrs) posted "Don't cry. Dance!" on Instagram — each screams into a wide-angle phone held inches from the face, they swap, then dance side by side to Bronski Beat's "Smalltown Boy".
The AI wave came on 2 October 2026, when @bellvtrix.ai posted a version with two cats made with Higgsfield Genjutsu, which passed a million likes. Since then people recast the dance with friends, partners, parents and characters, using AI video generators that keep the original camera and cuts and swap the people.
The song: "Smalltown Boy" by Bronski Beat
"Smalltown Boy" is the 1984 debut single of the British synth-pop trio Bronski Beat, from the album The Age of Consent, sung in Jimmy Somerville's falsetto. It reached the UK top three; its story is about a young gay man leaving a hometown that doesn't accept him.
Not the 2024 parents' dance or Just Dance
In 2024 "Smalltown Boy" also backed a different TikTok trend, "Dance like it's the 80s", where parents showed their old dance moves. Just Dance has its own routine too. Neither is this one: this page makes the 2026 scream-swap-and-dance duo.
Sources: Original post (Instagram) · Two-cat version (Instagram) · "Smalltown Boy" (Wikipedia) · 2024 "80s dance" trend (Know Your Meme)
TryH3 is an independent tool. It is not affiliated with, endorsed by or sponsored by Bronski Beat, its members or labels, the original video's creators or Higgsfield.
Smalltown Boy AI generators compared
What each dedicated Smalltown Boy page offered when we checked on 10 October 2026. Each site prices in its own credits, so compare the dollar cost of the pack you would buy.
| Tool | Inputs | Output | Song | Price per video |
|---|---|---|---|---|
| TryH3 | 2 photos (character sheet made from each) | 5–24 s, 9:16, 480p / 720p / 1080p | Included | From 14 credits; full 24 s: 245 / 542 / 954 credits |
| smalltownboydance.org | 2 photos (+ optional full-body), people or pets | 22 s, 9:16, 1080p | Silent — add it when posting | $9.99 one video, $24.99 three |
| FIKKU | 2 photos | About 24 s, 9:16 | Included | 45 credits |
| Flowlio AI | 2 photos | 21 s, 3:4, 480p / 720p / 1080p | Included | 500 (trend sale) / 1,600 / 2,800 credits |
| APT Video AI | 2 photos | About 21 s, 3:4, 480p / 720p / 1080p | Included | 15 / 35 / 60 credits |
TryH3 is the one that shows its models (Seedance 2.5 for the full 24 s, Seedance 2.0 for shorter openings), builds a character sheet from each photo before rendering, and keeps the reference's vertical cut with the song.
Best photos for the Smalltown Boy dance
The close-ups put each face a hand's width from the lens, so the photo matters more here than in most templates.
- One person per photo. Group shots and cropped strangers confuse who plays which role.
- Face the camera. A clear, well-lit, front-facing face survives the wide-angle distortion best.
- Show your hair or head. Hats, hoods and sunglasses carry into every shot, so leave them off unless you want them in the video.
- Waist-up or full body. The sheet copies the outfit you can see; a selfie gets a plain matching outfit below the shoulders.
- Skip heavy filters. Beauty filters and tiny, blurry crops make the face drift.
- Only people who agreed. Use photos of yourself and friends who are happy to be in it — no celebrities.
Captions to post with it
Tap one to copy it with the trend's hashtags, then post with the video.
Is the Smalltown Boy AI video generator free?
Signing up and uploading are free; each video costs credits — from 14 — and the price is shown before you generate.
Free
- Creating an account, uploading photos and playing the reference clip
- Generating on any account with enough credits — no plan or subscription required
- No prompt, editing app or filming needed
Paid
- Each video costs credits, shown next to the Generate button.
- The cheapest video is 14 credits (Seedance 2.0 Mini, 5 s, 480p), so the $5 Trial pack (24 credits) covers one short video. The full 24 s on Seedance 2.5 starts at 245 credits; the Creator pack ($49.90, 1,280 credits, no subscription) covers five.
- A Smalltown Boy video costs 14 to 954 credits, depending on the model, length and resolution.
- Videos made before your first purchase carry a moving tryh3.com watermark. After any credit pack, new videos have none, and saved watermark-free copies of earlier ones unlock.
- Credits are charged when you start and refunded automatically if a generation fails or a photo is not accepted.
Smalltown Boy AI video FAQ
Where can I make the Smalltown Boy AI video with my own photos?
On this page. Upload two photos, one person each, and TryH3 generates the scream-and-swap dance from a fixed motion reference, so nobody has to film it: 5, 10 or 15 seconds on any Seedance model, or the full 24 seconds on Seedance 2.5. The MP4 is vertical 9:16 and includes the Smalltown Boy track.
How many photos do I need to make the Smalltown Boy dance with a friend?
Two photos in total: one of you and one of your friend. You don't upload extra angles — TryH3 makes a front, back and profile character sheet from each photo itself.
Which photo screams first, and who dances on the left?
Photo 1 screams into the lens first (0–7 seconds) and stands on the left in the dance. Photo 2 takes the second close-up (7–14 seconds) and stands on the right. Upload order decides the roles, whatever the genders.
Can I use photos of two guys, two girls or a couple — or change the background?
Yes. Any two adults work. The reference shows a man and a woman, but roles follow the photo order, not gender, and each person keeps their own face, hair and clothes. You can also move the dance out of the courtyard: pick a red fence garden or a living room, or upload your own background.
What prompt do I use for the Smalltown Boy AI dance?
None. The camera, cuts and choreography come from the fixed motion reference, and TryH3 writes the instructions for the model. You only choose the photos and the resolution.
How much does a Smalltown Boy AI video cost?
From 14 credits (Seedance 2.0 Mini, 5 s, 480p). The full 24-second routine runs on Seedance 2.5: 245 credits at 480p, 542 at 720p or 954 at 1080p, both character sheets included. At the Creator pack rate ($49.90 for 1,280 credits) those are about $9.55, $21.13 or $37.19 of credits — credit costs, not single-video prices.
Is there a free Smalltown Boy AI video generator?
Signing up is free, and the $5 Trial pack (24 credits) covers a short 480p Seedance 2.0 Mini video (5 s for 14 credits, 10 s for 19); the full 24 seconds needs 245 credits on Seedance 2.5. Videos made before your first purchase carry a moving tryh3.com watermark.
Can I make the Smalltown Boy AI video without a watermark?
Yes, after any purchase. Videos made on an account that has never bought credits carry a moving tryh3.com mark; once you buy any pack, new videos have none and saved clean copies of earlier ones unlock.
What AI model makes the Smalltown Boy video? Is it Higgsfield Genjutsu?
The viral two-cat version was made with Higgsfield Genjutsu. TryH3 uses Seedance in reference mode instead — Seedance 2.5 for the full 24 seconds, Seedance 2.0 Mini, Fast or Pro for the 5–15 s opening — after turning each photo into a character sheet, so faces, hair and outfits stay consistent across the close-ups and wide shots.
Does the Smalltown Boy AI video come with the original Bronski Beat song?
Yes. The reference clip's Smalltown Boy track is laid under the finished video. The recording is copyrighted, so follow each platform's rules when posting, or mute it and add the song from TikTok's or Instagram's sound library.
Can I make the two-cat or dog version of the Smalltown Boy dance?
Not in this version. The template is built and tested for two people; pets aren't supported yet.
Will the AI keep our faces, hairstyles and outfits the same through the dance?
That's what the character sheets are for: the photo sets the face and the sheet sets the body, outfit and back view, so bald stays bald and long hair stays long. AI results still vary, especially in the wide-angle close-ups; a failed generation is refunded.
How long is the video, and can I get 1080p?
5, 10 or 15 seconds (the opening) on any Seedance model, or the full 24 seconds on Seedance 2.5. Every video is a vertical 9:16 MP4: 480p (480×854, the default) or 720p (720×1280), and 1080p (1080×1920) on Seedance 2.0 Pro and 2.5.
Where did the Smalltown Boy dance trend come from?
From a real Instagram video posted on 4 August 2026 by @elvira.mua.music and @mr_unvrs. The AI version took off on 2 October 2026 with a two-cat video by @bellvtrix.ai, made with Higgsfield Genjutsu.
What happens if my photos are rejected or the video fails?
On Seedance 2.5 the provider first reviews the sheets, the photos and the reference. If a photo is rejected or any generation fails, every credit comes back automatically. Only upload photos you have permission to use; how photos are stored is described in our privacy policy.
Make your Smalltown Boy video
Two photos, up to 24 seconds, the song included — see the price before you generate.
Upload two photos