
Full take · 23s
The full take runs to the rooftop
All 23 seconds: everything the short cut has, then the sprinter catching his breath, the waiting man laughing, the wide shot in front of the mosaic building and the closing stare into the lens.
Your two photos in the Ne-Yo closer dance AI meme.
Example · Full take · 23sPNG / JPG / WebP · up to 10 MB · at least 512px on the short side
0/2 photos
Photo 1 waits, photo 2 sprints in. One person or pet each, face clear; the thin glasses belong to the scene.
Video settings
Your own photos, a friend's or a pet's; anyone in them should know the video is being made.
The come closer dance meme is Gnome Boys' deadpan two-man face-off to Ne-Yo's "Closer": one of them waits in an empty car park and takes his glasses off, the other sprints in from the far end, they square up, grab at each other, and end up dancing side by side.

The joke is how serious they are: the two stare each other down as if it is about to turn into a fight, then spend the rest of the take dancing. The captions carry the other half — "We just can't stop", "we still can't stop", "who's gonna stop us" — a play on the "I just can't stop" in the chorus. It needs two people and an empty stretch of car park and nothing else, which is why friends, couples, families and whole teams keep refilming it.
This template uses the take from 2 April 2026, the one the current wave copies: the garage colonnade, the sprint in, the grabs under the beams and the finish on the roof in front of the mosaic building.
The AI layer keeps the cuts and only swaps the two dancers: the same edit turned up with a SpongeBob and Squidward cosplay, then TV characters and, days later, a chihuahua and a kitten; mini-tool sites and app templates followed within two weeks, all of them built on “swap the people in the clip”. That is what this template does with two photos.
Ne-Yo releases "Closer", the song the dance is performed to.
Gnome Boys post the first take, "We just can't stop", on a tree-lined path: about 14.9 million plays on TikTok when read on 3 November 2026.
A second take, "we still can't stop", on a park path: about 15.1 million plays.
The third take, "who's gonna stop us", moves to a parking garage and its roof: about 12.5 million plays, and the remakes that follow copy this cut.
Three TikTok accounts post the same "how to make AI come closer dance" walkthrough on one day, after the first AI character version on 7 October: the AI layer and its name arrive.
Year of the record from Ne-Yo's release listings; account names, post dates and play counts read on TikTok on 3 November 2026.
The whole 23-second take in eight stops, each one with the second its shot starts on.
10:00–0:01The camera sits at the waiting man's shoulder while the other one jogs the length of the colonnade, small in the distance.
20:01–0:02He takes his glasses off, drops them out of frame and stares down the lens without a word.
30:02–0:04The sprinter comes at the camera flat out, flashes past a pillar and lands a leaping step into frame.
40:04–0:09They meet under the beams, already bouncing, and copy each other's shoulder rolls with the city behind them.
50:09–0:11They grab at each other's jaw and wrists, nose to nose, and neither one steps back.
60:11–0:15They split apart into the loosest part of the routine, spinning away and back in the open level; this is where the short cut ends.
70:15–0:17A close-up of the sprinter catching his breath, then the waiting man looking up and laughing.
80:17–0:22Side by side in front of the mosaic building they dance to the end, then the last shot lands on the waiting man staring into the lens.
Both cuts take the same two photos and the same settings, so the only difference is how much of the take you keep. The short cut is the cheaper run.

Full take · 23s
All 23 seconds: everything the short cut has, then the sprinter catching his breath, the waiting man laughing, the wide shot in front of the mosaic building and the closing stare into the lens.

Short cut · 15s
15 seconds: the wait, the glasses, the sprint, the face-off, the grabs and the two of them dancing in the open. It stops on the wide shot of the two of them, on the frame before the close-up of the sprinter comes in.
Four different things turn up for these keywords; only the first one is the Gnome Boys face-off that this template recuts with your photos.
| Compare | Come closer dance | Wizkid's Come Closer | Other Closer songs | Godzilla x Kong edits |
|---|---|---|---|---|
| What it is | The Gnome Boys face-off | A 2026 dance challenge | Same title, other songs | Fight scenes cut to it |
| The record | Ne-Yo ‘Closer’, 2008 | Wizkid ft. Drake, 2017 | Tyla, SOMI, Chainsmokers | Ne-Yo ‘Closer’ on clips |
| The move | Wait, sprint, square up | A pull-them-closer hand | Nothing shared | Trading blows and beams |
| What you can make here | Your two photos in it | Nothing; another dance | Nothing; another song | Nothing; other footage |
Each photo plays one of the two dancers: photo 1 waits in the garage, photo 2 sprints in. A friend, a partner or a pet can take either part, and swapping the two photos swaps the roles.

Photo 1·Waits
Face, hair and build for the man who waits, takes his glasses off and stares; he is in the first shot and in every one after it.

Photo 2·Sprints in
Face, hair and outfit for the man who runs the length of the garage, grabs at the other one and finishes side by side with him.
Works well
One person, face clear
Head to waist or full body
A pet can take a partAvoid
Two people in one of the two photos
A hand or a cap over the face
Blurry, dark or filteredAdd the two dancers
Photo 1 waits, photo 2 sprints in; the short cut and the full take both use both.
Click Generate
The button shows the exact price for the cut and the resolution you pick before you start.
Post it
The download is already vertical and ready for TikTok, Reels or Shorts.
This is a two-person trend: it needs two people and nothing else — no rehearsal, no car park of your own, no third friend holding the phone.
The default way to use it: one of you waits with the glasses, the other sprints in.
Example: DC: the friend who is always late, now sprinting
Both parts take anyone, so a couple can split the two photos however they like.
Example: the couple that argues about everything, now arguing in a car park
Two brothers, two sisters, or a parent and a grown-up child each take one photo.
Example: sibling rivalry, with a garage for a venue
Pets can take either part; they run in, face off and jump up on their hind legs.
Example: the cat waits, the dog sprints in
One copyable prompt for both cuts, written for video-edit models: attach the take as the clip to edit, add the two photos, and the same wording also works when the clip goes in as a reference in Seedance or MiniMax. The prompt this page runs is already set: add two photos and generate.
Video edit
Prompt length 1173
@Video1 is the garage take, and @Image1 and @Image2 are the two people who take over its two parts.
Put the person in @Image1 on the man who waits in the colonnade, the one in the black sweater with the thin clear glasses: he is in the first shot and in every shot after it, close-ups included. Put the person in @Image2 on the man who jogs the length of the garage, sprints past the pillar, grabs at the other one and dances beside him at the end.
Both of them repeat the performance of @Video1 exactly as it plays: the glasses coming off and dropping out of frame, the stare into the lens, the sprint in, the shoulder rolls, the hands on jaw and wrist, and the last stare on the roof. Same timing, same order, beat by beat.
Nothing else about the take changes: the car park, the colonnade and the rooftop stay as they are, the camera keeps its framing and its moves, the cuts land where they land now, and the light and the pace are what they already are. The clip's own soundtrack stays with it.
Two faces, two bodies: each new face keeps its own hair and clothes and never blends with the other face or with the face it replaces. No extra people, no text, no logos.One face per photo
Each photo plays one dancer, so a single person or pet per upload is what keeps the two parts apart.
Leave the glasses to the clip
He takes them off in the first close-up; they belong to @Video1, so the prompt never has to add or remove them.
Say what stays
Naming the sprint, the grabs, the cuts and the track is what keeps it the same video with a new cast.
“Use one photo for both parts”
One photo on both dancers is the fastest way to one shared face on two bodies, and the face-off is over.
“Change the song to …”
A video edit has no audio field and keeps the clip's own track; asking for music makes the model write a new one.
“Cut it to 15 seconds”
Length comes from the clip you attach, not from the words; both cuts take the same prompt.
Describes a trend, not a clip
Put my two photos in the come closer dance video and make us do the dance.
Names the clip and what stays
Edit @Video1: @Image1 takes the man who waits, @Image2 the one who sprints in; keep the cuts and the clip's own sound.
The model has never seen the trend; it only knows the clip you attach and the words that say who replaces whom and what stays.
Every template works the same way: add your photos, press Generate and get a shareable video without writing a prompt.
View all effectsOne photo, and the dance everyone is copying.
Try itOne photo shows the gear to the camera.
Try itOne photo puts you at the centre of the playground crowd in Bakar's Lonyo! music video, school uniform and song included.
Try itYour pets or you take over the kitchen coffee clip, milk spill and all.
Try itYour cat or dog walks up on two legs and flips two middle fingers.
Try itOne photo flexes the bicep heart at the camera.
Try itOne photo mixes the $1, $5 and $50 drinks.
Try itThe take arrives cut and finished: these tools upscale it, extend it, or put the same dance on a different photo.
Generate a clip from a prompt, a still image, a reference or existing footage.
Open toolFAQ
No, each run costs credits at the price of the cut and the resolution you pick, and Generate shows the exact number first. New accounts start with 10 credits.
Yes, the result keeps the soundtrack the take already carries; there is no field to change it, so the dance and the music stay in step.
The dance is performed to "Closer", from Ne-Yo's 2008 album Year of the Gentleman. The take carries its own soundtrack, and the result keeps that.
Gnome Boys, the Australian duo behind @gnomeboys, posted the first version on 10 March 2022 as "We just can't stop"; they refilmed it in December 2023 and again on 2 April 2026, and that third take is the one this template recuts.
No. Wizkid's "Come Closer" with Drake is a different song with its own 2026 hand-gesture challenge; this one is Ne-Yo's "Closer" and the Gnome Boys face-off.
No. Both photos go in from the browser and the finished video comes back, so there is no app to install and nothing to edit by hand.
No lessons needed. Every step — the wait, the sprint in, the grabs and the dance at the end — comes with the template, so nobody has to learn the routine or rehearse before you make it.
Yes. Each photo takes one of the two parts, so two friends, two sisters or a couple of any mix work; nothing in the take assumes two men.
Yes, a pet can take either photo. Running in, squaring up and jumping on the hind legs is the hardest thing this template asks of a cat or a dog, so expect a loose frame or two; keep the whole animal and its face in the shot.
The take cuts about twenty times and several shots put the face much further from the lens than the opening close-up, so those frames get fewer pixels. A sharper, evenly lit face photo is what fixes it, and the checklist above says what else helps.
No. This page recuts a take that Gnome Boys posted, and it has no connection to them or to Ne-Yo.
Videos made on a free account are public. On a paid plan they are private by default.
Come Closer Dance AI
Add two photos, pick the short cut or the full take and generate: the garage, the rooftop and every cut are already in the template.