Guide
Two people dancing, from two photos
Two photos in, eight seconds of two people dancing out. The idea is simple; the part that decides whether the clip is usable is the photos you pick and the hands the model has to invent. This page covers both, plus what it costs and what to do when the first render misses.
What it actually does
You give the model two still photos. It keeps the faces and the clothes, invents the room, and generates eight seconds of movement in which both people perform the same short routine. Nothing is lip-synced, nothing is spoken, and the clip is silent: it is movement, not performance.
On this site the routine is fixed — the hop hop hop dance, the one with the point, the three hops, the flat palms and the hard cut. That is deliberate. A single format is easier to get right than a menu of dances, and the price stays on the button instead of hiding behind a model picker.
Duo or Twin: one photo or two
Duo takes two photos and keeps two different people in frame. This is the version the trend is built on: two friends, a couple, a parent and a grown child, a person and their dog.
Twin takes one photo and puts the same adult in both roles. It is the answer when the second person is asleep, away, or a cat that will not sit still. It costs less (30 credits against 40) because there is one face to keep consistent instead of two.
The photos that actually render
- One person per photo. A group photo makes the model guess which face belongs to which body, and it guesses wrong.
- Front-facing, eyes open, no sunglasses. A profile or a hidden face gives the model nothing to hold on to across the cut.
- Full body, feet visible. The hops are the middle of the clip. If the photo ends at the knees, the model invents legs.
- Even light on both photos. One person in daylight and one under a orange lamp makes the pair look pasted together.
- Sharp, not blurry. A screenshot of a story is usually softer than it looks, and soft faces drift during the render.
What breaks a render
Hands are the weak point: the palm beat is where a finger can merge or a thumb can appear twice. Feet are second. Mirrors are third, because the model can keep the reflection and forget the person.
None of this is a reason to give up on the first try. Generate once, look at the hands, and if they smear, try a photo where the hands are already visible and relaxed. A render that fails outright returns its credits automatically; a render that is clearly broken can be refunded by email.
What it costs, before you upload
The price is printed on the generate button: 40 credits for Duo, 30 for Twin. Credits come in three packs — 200 for $4.99, 600 for $11.99, 1,500 for $26.99 — and they never expire. A new account starts with 40 credits, which is one Duo clip, and a guest gets one free clip per device per day without signing in at all.
Payment runs through Waffo, which is the merchant of record: it takes the card, calculates the tax and issues the receipt. A failed render puts the credits back automatically.
What you get back
An MP4, eight seconds long, 9:16, 720p, no watermark. It downloads straight from the result card and stays in your account for 30 days. It is silent on purpose — the trending sound is licensed by the platform, so you add it inside TikTok, Reels or Shorts when you post.
Posting it without getting it wrong
- Label the post as AI-generated, using the platform's own label.
- Add the sound in the app, and trim it so the first beat lands on the point.
- Do not caption it as real footage of the people in the photos.
- Ask the second person before you post them, even if they agreed to the render.
Other dances, honestly
This site renders one format. If you want a different routine — a slow dance, a wedding first dance, a K-pop chorus — a general video model with a reference clip is the better tool, and it will cost more and need more attempts. We would rather say that than pretend a single prompt can do every dance well.
Questions about two-person clips
Can I make two people dance from photos?
Yes. Upload one photo per person, pick Duo, and the model renders both of them performing the same eight-second routine. The faces come from your photos; the movement is generated.
Is there a free version?
The first clip is free without an account, one per device per day. After that a Duo clip costs 40 credits, and the smallest pack is $4.99 for 200 credits — five clips.
Can the two people be any age?
No. Both must be adults who agreed to appear, and you confirm that before every render. Minors are not allowed, and neither are public figures — no celebrities, politicians or other well-known people, not even as a joke.
What if I only have one photo of the person?
Use Twin mode. It puts the same adult in both roles, so a single photo is enough — useful for a solo version of the trend, or when the second person is not available.
How long does it take?
One to three minutes for the render. Signed-in users get an email when it is done, and the clip stays in the account for 30 days.