Hotline Bling AI Video Generator

One photo starts the dance in the glowing pink room and takes both close-ups. A second photo hands the low crouch, the magenta wall and the green ending to somebody else.

Made from two photos of fictional adults.

Make yours

120 credits

Your photos

1 needed

Slot 1 is required and owns the opening room plus the two close-ups. Slot 2 is optional: fill it and that person gets the crouch, the purple wall and the green ending, leave it empty and the person in photo 1 dances both parts. One consenting adult per photo, face lit and clear, wearing what they should wear on screen.

Soundtrack

Keeps the clip’s own audio under your video.

What you get

Fixed

Length
13 s
Format
16:9 landscape
Sound
Clip audio
Ready in
A few minutes
Cost
120 credits 1 video

Sign in with Google at the next step.

Failed renders are refunded automatically. You can leave the page while it renders.

By making a video, you confirm everyone in the photos agreed to it. Terms

Short answer: Hotline Bling AI is a TikTok format that spread from November 5, 2026: an AI clip rebuilds the bare rooms of a 2015 music video, each one lit in a single strong colour, and walks two dancers through them a shot at a time. On Genvima you hand the effect a photo of yourself, plus a second photo if somebody is dancing with you, and it films those seven shots again with your own faces and clothes. What comes back is 13 seconds at 16:9, 480p MP4, carrying the posts' own audio unless you switch the sound off, finished in a few minutes and clean of any watermark. One render costs 120 credits, which the $9.99 credit pack pays for, and the effect has no free version. Genvima has no ties to Drake, to his record label, or to the people who built the template the trend runs on.

How to make a Hotline Bling AI video on Genvima

The maker sits at the top of this page. Nothing is typed: the shots, the rooms and the timing are already set, and your photos decide who dances them.

  1. 1

    Put your photo in slot 1

    This is the part with the most screen time: the small figure at the far wall of the pink room, then both close-ups, then the walk across the white room. Pick a well lit picture of one adult looking toward the lens, in whatever clothes you want to be seen in.

  2. 2

    Add a second person, or do not

    Slot 2 gives the crouch, the magenta wall and the green ending to a partner, a friend or a sibling. Leave it empty and the person from photo 1 covers all seven shots instead. Nobody is invented to fill the gap, and the 120 credits are the same either way.

  3. 3

    Keep the audio or mute it

    The default lays the audio from the trend posts under the picture. Choose the silent option when the app you post from supplies its own track, or when you plan to cut the clip into a longer edit.

  4. 4

    Press Remake this and come back later

    Google sign-in happens on the first render; the photos you already chose are held and the job starts on this page. The 13 second MP4 shows up under the maker and in Assets within a few minutes, and closing the tab does not stop it. Say it is AI-made when you post it.

What your photos decide, and what every render keeps

  • Slot 1 opens and owns the close-ups

    That face appears five times: tiny against the far wall in pink, twice filling the frame in the icy blue room, once with a hand cupped at the ear, and once stepping sideways through the white room. It is the part people will recognise first.

  • Slot 2 gets the three colour rooms

    The second person drops into a low crouch on the heels in pink, with the head turned over the shoulder at the camera, leans against a wall under purple light, and finishes the clip turning into the lens in green.

  • One person can hold the whole clip

    With a single photo the same dancer plays both halves, room by room, which is how plenty of the posts look. Genvima never invents a second guest to stand in, so you always know who is on screen.

  • Nobody is on screen with anybody else

    Each of the seven shots holds exactly one dancer. That is why two sharp solo photos beat one group picture, and why the pair never has to match heights or stand side by side.

  • Your clothes, your hair, your build

    Everyone arrives as they are, in the outfit their photo shows. No costume is added, no cap or sunglasses appears unless the photo already had them, and the dancers stay ordinary people without ears, tails or paint.

  • Light that paints everything it touches

    The rooms are empty and the colour is the whole set. Deep pink, pale blue-white, magenta and teal wash across skin, hair and fabric, so a white shirt reads pink in one shot and green in another. That shift is the look, not a fault in the render.

  • The cut list never moves

    Seven shots, switching on the beat, in the same order and at the same lengths every time. Camera heights, room colours and the point where each person comes and goes are fixed by the template.

  • Audio laid under a silent render

    The picture is generated without sound, then the 13 seconds of audio from the posts are muxed underneath. Mute it instead and you get a quiet file that any editor can drop a track onto.

The same effect with two photos and with one

Two renders exactly as they came out of the model, both built from photos of fictional adults.

  • Two photos: he takes the close-ups, she takes the colour rooms

  • One photo: she dances both parts in her black dress

Hotline Bling AI at a glance

Photos
1 to 2, one adult in each
Parts
Person A: pink wide shot, two close-ups, white room. Person B: the crouch, purple, green
Empty slot 2
Person A dances all seven shots
People per shot
One
Clothes
Whatever each photo shows
Clip
13 seconds, 16:9, 480p, MP4
Sound
Audio from the trend posts, or silent
Trend
On TikTok since November 5, 2026
Price
120 credits a render; the $9.99 pack buys one
Render time
A few minutes
Watermark
None
Made with
Seedance 2.0 Mini
Every value comes from the effect recipe, so the page cannot promise a length, a shape or a price the render does not deliver.

Who is posting this one

  • Couples

    The template was built for two people trading shots, and a couple is the obvious cast: one of you opens the clip and keeps both close-ups, while the other gets the crouch and the final room. It reads as a joke, not a love letter, which is why it travels.

  • Best friends and roommates

    Two photos from the camera roll, no plan, no rehearsal. The pair never share a frame, so you do not even have to be in the same city when you make it.

  • Siblings

    Family chats pass this one around because the moves are unmistakable and nobody had to learn them. Two childhood faces in grown-up clothes under magenta light does the rest.

  • Solo creators

    One photo and the whole clip is yours. Seven shots of the same person works like a short music video and is easy to caption as a mood rather than a duet.

  • Birthdays and anniversaries

    Put the person being celebrated in slot 1 so they open the clip and hold both close-ups, and give slot 2 to whoever is throwing the party.

  • Dance accounts and small brands

    The moves are loose and goofy on purpose, so it fits an account that posts trends rather than choreography. Ask anyone whose photo you use before you upload it.

Where the Hotline Bling AI trend came from

The reference everyone has in their head is Drake’s 2015 video for "Hotline Bling", shot in a tall empty studio where the singer danced by himself and the walls changed from pink to blue to green between cuts. It has been a meme almost continuously ever since.

The 2026 version is different in one important way: it is an AI template. From November 5 accounts began feeding cartoon and film characters into a clip that copies the look of those rooms and gives two dancers a shot each, and the same cuts went around again on November 9 and November 10 with new pairs of characters in them.

Those characters are somebody else’s designs, and the music video belongs to its label, so this page is not a way to get either. The effect here re-films the clip from scratch with real adults from your photos: the rooms, the poses, the camera heights and the timing are rebuilt, and not one of those characters, no singer and no footage from the 2015 video ends up in your file. Genvima is independent of Drake, of his record company, and of whoever assembled the template the posts use.

The thirteen seconds, shot by shot

Seven shots, one dancer in each, switching on the beat. A is the photo in slot 1 and B the photo in slot 2.

  • 0 to 2.5 s, pink, wide and low: A stands small against the far wall of a glowing pink room, bouncing through a loose dance that fills the whole body.
  • 2.5 to 4.5 s, icy blue, close: the camera drops low and A fills the frame, arms swinging out wide, eyes half shut.
  • 4.5 to 6 s, pink again: B is down in a crouch on the heels, side on to the lens, head turned round toward the camera.
  • 6 to 7.8 s, icy blue, close: A once more, one hand cupped at the ear like a telephone, the other waving loosely.
  • 7.8 to 9.3 s, purple: B leans into a magenta wall and rolls the hips without going anywhere.
  • 9.3 to 11.5 s, white: the brightest shot in the clip. A, seen wide, steps side to side with a hand still raised at the ear.
  • 11.5 to 13 s, teal green: B sways and turns to face the camera as the clip ends.

Count it up and slot 1 carries roughly eight of the thirteen seconds, including every close-up, so that photo deserves the better picture of the two.

Which photos work and which ones waste a render

  • One adult per picture. Crop a group shot down to a single person before you upload it. Two faces in one photo confuses which of them should dance.
  • Dressed as you want to appear. Clothes are copied, not invented, so choose the picture for the outfit as much as for the face. Remember the coloured light will tint it.
  • Bright, even light on the face. The close-ups are tight, and a dim or motion-blurred photo shows there more than anywhere else.
  • No sunglasses, no hats you do not want. Accessories come along when the photo has them and stay off when it does not.
  • Half body or more beats a head crop. When the legs are outside the frame the model picks plain trousers and shoes to match the top, which may not be what you own.
  • Adults who agreed, and nobody famous. Pictures of public figures and of children are turned down, and so are drawings, characters and animals.

One photo or two

  • One photo: the person in slot 1 dances every shot, pink through green. This is the quickest version and it is what most single-person posts look like.
  • Two photos: slot 1 opens the clip and takes both close-ups and the white room, slot 2 takes the crouch, the purple wall and the green ending. They never appear together in a frame, which is exactly how the template was built.

Both cost 120 credits. If you want more people in one clip, No Scrubs AI puts a trio in a row and Tiny Desk AI casts a whole band; every other model and effect is listed on the AI video maker page.

How close the render gets to the posts

Kept: the seven shots and their order, the lengths, the room colours, the camera heights, the poses, and the one-person-per-shot rule.

From your photos: the two dancers, their faces, hair, builds and clothes, and whether one person or two appear.

Different on purpose: the characters are gone. Our dancers are ordinary adults with no tails, ears or face paint, because a re-film with real people is the only version we will sell.

Different because models are imperfect: the rooms are plain painted boxes with no furniture or lettering, the colour wash can run a shade warmer or cooler than the posts, and a fast arm sometimes smears for a frame or two.

Why the clip is landscape

16:9 is the shape the AI posts use, borrowed from the music video they imitate, and the wide rooms are the point: crop it to a phone frame and the glowing walls on either side of the dancer disappear.

On TikTok, Reels and Shorts the picture sits in a band across the centre of the phone, dark above it and dark below, just as the trend posts do. Many accounts use that empty space for a caption naming who is who. If an app replaces the audio, render with the sound off and add a track from the app’s own library.

Price, time and the file you get

One Hotline Bling AI render is 120 credits whether you upload one photo or two. The $9.99 credit pack covers a single video, and the plans and larger packs on the pricing page bring the cost of each one down. There is no free version, because the effect needs your photos to work.

The job runs on Seedance 2.0 Mini and finishes in a few minutes. You get a 13 second 16:9 MP4 at 480p with no watermark and no logo, stored in Assets and downloadable as often as you like. If a model fails or refuses the request, the credits for that clip go back to your balance automatically.

To write your own scene instead of dancing this one, the same model is open as a plain generator on the Seedance 2.0 page.

FAQ

What exactly is the Hotline Bling AI trend?

It is a TikTok format that took off from November 5, 2026. An AI template recreates the bare, single-colour rooms of a 2015 music video and sends two dancers through seven shots, one at a time; the first wave of posts filled it with cartoon and film characters. Genvima’s effect films the same clip with adults from your photos instead.

How do I get my own face into a Hotline Bling AI video?

Open the maker at the top of this page, drop a photo of yourself in slot 1, add a second person in slot 2 if you want one, keep or mute the audio, and press Remake this. A 13 second MP4 is ready a few minutes later in Assets.

Do I need one photo or two?

One is enough. Slot 1 is the only required photo, and with slot 2 empty that person dances all seven shots. Fill slot 2 and the two of you split the clip. The price does not change.

Which shots am I in?

The person in slot 1 starts as the small figure at the back of the pink room, fills the frame in both icy blue close-ups, lifts a hand to the ear and crosses the bright white room. Slot 2 is down on the heels in pink, leans under magenta light and ends the clip in green.

Can one person be both dancers?

Yes, and that is what a single photo gives you: the same face in every shot. There is no version where the model copies your face onto an invented partner.

What will we be wearing?

The clothes in your photos, shot for shot. No outfit is added and no costume is put on anyone. Keep in mind that saturated pink, purple and green light falls on whatever you wear, so colours read differently from room to room.

Is Drake or the real 2015 video in my clip?

No. Nothing from the music video is in the file: every frame is newly generated around the people in your photos, and the clip only follows the kind of rooms, the camera heights and the cuts that the AI posts use. Genvima has no connection to Drake or his label, and photos of famous people are refused.

Can I use a cartoon character like the posts do?

No. The template the trend started with used copyrighted character designs, and we do not reproduce those. This effect takes photos of real adults only, so drawings, game or film characters and anime art are turned down.

Can a pet or a drawing be one of the dancers?

No. The shots are built around human bodies doing human moves, so animals, illustrations and 3D renders are all turned down. People only, one adult in each photo.

Can I upload one photo of the two of us?

No. Each slot takes one person on their own, which keeps both faces sharp and puts each of them in the right shots. Crop the two of you into separate pictures and upload them one per slot.

Is the song on the video?

With the sound kept on, the file carries the 13 seconds of audio from the trend posts. With it off you get a silent MP4, which is the better choice if you plan to pick a track inside TikTok or Instagram.

Why is the video landscape and not vertical?

The AI posts are 16:9, like the video they copy, and the wide empty rooms are half the effect. On a phone it fills a strip through the centre of the screen, with black above and black below.

How much does a Hotline Bling AI video cost?

120 credits a render, with one photo or two. The $9.99 credit pack is exactly one video; monthly plans and bigger packs on the [pricing](/pricing) page work out cheaper per clip.

How long does the render take?

A few minutes for the 13 second clip. Leaving the page is fine: the finished video waits for you in Assets.

Is there a watermark on the file?

No. The MP4 comes out clean, with nothing written over the picture.

Why did my Hotline Bling AI video fail?

Usually the model’s safety check stopped a photo before filming began. It sometimes decides a picture shows a well known real person even when it does not, and then nothing is rendered. Your 120 credits go straight back to your balance. Try the same person in a different photo: a plain background, even light, one face in the frame.

Are photos of celebrities or children allowed?

No. Every person in the clip has to be an adult who agreed to appear in it. Pictures of public figures and of minors are refused.

Is Hotline Bling AI free on Genvima?

No. There is no free version of the Hotline Bling AI effect: it works from your photos on a paid model. Genvima's free Basic engine (one free video and three free images a day) makes clips and pictures from a written prompt only, so it cannot use your photos. One render is 120 credits.

What happens if the render fails?

If a model fails or refuses the request, the credits for that clip go back to your balance automatically.

Keep exploring

Try Hotline Bling AI now

Add your photo and see the price before you generate.

Back to the top