Add the left photo, then the right photo. The order decides who is who in the scene.
What this template is
Neon Shuffle puts two people side by side in one vertical frame, dancing a sharp, mirrored shuffle in a dark room lit by cyan and magenta. Both people move at the same time, which is why the template needs two photos: there are two bodies on screen for the whole clip and no moment when one of them is out of shot.
The template also decides who stands where. The first photo becomes the dancer on the left, the second becomes the dancer on the right. Swap the order and the two people swap places in the clip, so the order is a creative choice, not a formality.
This is the template to pick for a couple, two friends, two siblings or a band: two people in one clip, each keeping their own face and clothes, in a scene neither of them has to be in to film.
How the reference clip is used
The model copies the movement and camera work from the template's own clip and swaps in your face. That is why every result keeps the same rhythm as the template, and why you never write a prompt.
What the video looks like
Each pair is one upload: the photo on the left, the clip it turns into on the right.
Result
6 s
Two people, short cut
Two separate photos become one pair: the first on the left, the second on the right, in the same neon room.
Result
10 s
Two people, full cut
The same setup at the longer length, with more of the shuffle before the last beat settles.
The people in these clips are placeholders made for this demo, not real users.
Both of you in one clip without a tripod, a shoot or a second take: two photos and one render.
A friends or band reveal
Put two bandmates or two friends side by side as a lineup, each one recognisable on their own side.
A birthday shout-out
Two faces, one clip, sent to a group chat with a caption that names the occasion.
A long-distance clip
Two people who are never in the same room get one frame together without booking a flight.
Which photos to upload
Two photos, and each one has its own job in the clip:
1
Left
The person on the left of the frame. Their face, hair and outfit are copied onto the left dancer.
2
Right
The person on the right of the frame. Their face, hair and outfit are copied onto the right dancer.
Two separate photos, one person in each, both facing the camera with the head and shoulders in frame.
Matching light in both photos. One photo in daylight and one in a dark room still works, but even light on both faces gives a more convincing pair.
Photos where the outfit is visible. A jacket or a distinctive top is what makes each person recognisable in the wide part of the frame.
Two people of any age above 18, in any combination: partners, friends, bandmates, relatives.
One group photo used twice, or the same photo in both slots. The model needs two different faces to keep them apart.
A photo where the person is turned away or partly out of frame; the face is what the model copies.
Heavy filters or face-swap apps applied before upload. Each one gives the model a face that is no longer the real one.
Sunglasses, masks or a hat pulled down over the eyes, for the same reason as on the one-person template.
Step by step
1
Add both photos
Left and right are different people. Each face should be clear and facing the camera.
2
Pick a length
6 or 10 seconds. A longer clip costs more credits.
3
Download
Both people stay in their own clothes. The result is AI-generated.
How the clip plays out
The clip runs 6 to 10 seconds, and the template fixes the camera and the pacing; you only swap in the person.
Both people are in frame from the first second, and the template keeps them there: no cut away, no solo close-up. The early part of the clip holds the pair in a wide two-shot, the shuffle builds in the middle, and the last beat settles both of them into a matching pose. The exact framing comes from the template clip, so treat this as the shape of the clip rather than a shot-by-shot script.
How it differs from the one-photo templates
Soft Bounce and Glow Poster take one photo each; this one needs two, and that is the only real difference in how you use it. Nothing else changes: same generator, same one-tap render, same fixed prompt.
The price is higher than the one-photo templates because the render is a different setup, not because you upload more files. One render still produces one clip.
Lengths and credits
Length
Credits
Good for
6 s
10
A short version of the duo clip. Good for a quick post.
10 s
18
The full two-person clip, with more room for the shuffle.
2 photos720p9:16From 10 credits
Two photos do not cost twice as much: one render produces one clip, and the price depends on its length. The short version starts at 10 credits, and the full version costs more because it renders more seconds. The Generate button shows the exact cost before you start, and a failed render returns its credits automatically.
The prompt behind this template
Replace the two people in @Video1 with @Element1 and @Element2, keeping the exact same movement, camera path and lighting. They dance a sharp neon shuffle side by side in a dark club with cyan and magenta light. @Element1 is the person on the left, @Element2 is the person on the right. Keep each face, hair and outfit. One continuous vertical shot, no text, no watermark.
This wording is fixed on the template. You never have to write it yourself.
Model: Kling 3.0 Omni
When it comes out wrong
Problem
Fix
The two people are on the wrong sides
The first photo is always the left dancer. Swap the two photos and render again.
The two faces blend together
Use two photos with one person each. A group photo, or the same photo in both slots, makes the model guess.
One person looks less like their photo
Use a sharper, front-facing photo for that slot. In the wide two-shot each face is small; the sharper file holds up better.
The clothes came out mixed
The model reads each outfit from its own photo. Clear, contrasting outfits keep the two people apart.
The render was refused
The safety check rejects some photos. Credits are refunded; try a different photo for the slot that failed.
Posting it
Name both people in the caption, or tag them, so viewers know who is who before the clip starts.
Add a sound with a clear drop, because the clip has no audio of its own and the shuffle lands on the beat.
Keep the AI label on, and mention that the video is AI-generated if you post it without the platform label.
Ask both people before you post. A clip of two people needs two yeses, not one.
Questions about this template
How many photos does Neon Shuffle need?
Two, one person in each. The first photo becomes the dancer on the left, the second the dancer on the right. Two photos of the same person also work if you want the same face on both sides.
Can I use two pets instead of two people?
The template is built for people: the prompt asks for a face, hair and an outfit. A clear photo of a pet usually comes back unrecognisable or partly human, so it is not the template to pick for animals.
What if the two people are different heights?
The template fits both people into the same two-shot, so the clip follows its own framing rather than your photos. A big height difference still shows, just not as a literal measurement.
Does the order of the photos matter?
Yes. The first slot is the left of the frame. If the result has the two people swapped, swap the photos and render again: nothing else changes.
Do both people need to consent?
Yes. Only upload photos of yourself and of adults who have given you permission. Never upload a photo of a child, a celebrity or a public figure.
How long does the video take, and does it have sound?
A few minutes, usually. The clip comes back silent, so you add a sound when you post it.
