Best friends
The classic version. Dress the way you would for a night out — the clothes in your photos are the clothes in the video.
The rap duo trend, from two photos
Add a photo of you and one of a friend. The AI puts the two of you into the Hotel Lobby performance — the same moves, camera cuts and audio as the original — lip-synced in the orange booth, ready to post.
Photo 1 · left side
Photo 2 · right side
4:3 768p MP4 with the original performance audio. Failed renders are refunded automatically.
Made with this generator
Made from two photos of AI-generated people — they are not real. Examples on this page play without sound; your download includes the original performance audio.
The Hotel Lobby trend recreates “Hotel Lobby”, the 2022 performance by Quavo and Takeoff — uncle and nephew, two of the three members of Migos, who also released music as Unc & Phew — filmed for A COLORS SHOW in a plain orange booth with a single hanging microphone.
In September 2026 people started using AI to put themselves, their friends, their pets and celebrities into that booth. The most-viewed versions keep the original moves and audio and only change who is performing. The format spread on TikTok and Instagram, where you will also see it called the Migos AI trend or the rap duo trend.
This generator makes that version from two photos: a 12- or 15-second cut of the performance, with the two of you in place of the original performers. It is not affiliated with Quavo, Takeoff, Migos or COLORS.
There is no single official filter. CapCut templates paste a face onto a fixed clip. This generator re-renders the whole performance from your two photos, so your faces, hair and clothes move with the body and your lips follow the original verse.
One adult per photo, face clear and front-facing; waist-up or full-length shows your outfit. Photo 1 takes the left spot, Photo 2 the right.
12 seconds covers the back-and-forth in the wide shot; 15 seconds adds the close-up at the mic. 35 or 44 credits.
In about 7–12 minutes the video is in your Library. Download it with sound, or the silent copy to add the Hotel Lobby sound inside TikTok.
The performance is a rap duo, so it always takes two. Pairs that work well:
The classic version. Dress the way you would for a night out — the clothes in your photos are the clothes in the video.
Matching or clashing outfits both read well against the orange booth.
Siblings, a parent and a grown-up kid, a grandparent and an adult grandchild — the uncle-and-nephew original started it.
A farewell or launch video the whole team will share.
People make Hotel Lobby AI videos in three main ways. They differ in what you start from, how much of the original performance you get and what you hear.
| Way to make it | You start from | What you get | Audio |
|---|---|---|---|
| CapCut template | A ready-made template clip | Your face placed onto a fixed clip | Usually added by you in the app |
| Motion-transfer apps (e.g. Higgsfield) | The original performance video plus your photos | The original choreography, re-rendered with you | Often the original track |
| This Hotel Lobby AI generator | Two photos — no video to find or upload | A 12- or 15-second cut of the performance, re-rendered with both of you and lip-synced | The original performance audio, plus a silent copy |
Other apps change their features and prices often, so check them before you pay. Videos that contain the original recording — including ones made here — can be muted by copyright checks on some platforms.
Results vary: occasionally a face drifts, a gesture comes out smaller than in the original, or a piece of clothing changes. Failed renders are refunded automatically. Videos are rendered by the MiniMax H3 model.
Pricing, before you click
A 12-second video uses 35 credits and a 15-second video uses 44, from any plan or credit pack. New accounts get 10 credits, which is not enough for a video. Finished videos download without a watermark.
No. A 12-second Hotel Lobby AI video uses 35 credits and a 15-second one uses 44. New accounts get 10 credits, which is not enough for a video. If a render fails, the credits go back automatically.
Add a front-facing photo of each person, choose 12 or 15 seconds and press generate. About seven to twelve minutes later you get the Hotel Lobby performance with the two of you in it.
Neither. A CapCut template pastes a face onto a fixed clip; this re-renders the whole performance from your two photos, so your faces, hair and clothes move naturally. You can still finish the video in CapCut or TikTok afterwards.
Yes. The video keeps the audio of the original COLORS performance, and your lips are synced to it. Some platforms mute videos that contain the original recording; if that happens, download the silent copy and add the Hotel Lobby sound from the platform’s library.
Quavo’s and Takeoff’s, from the original performance. The video does not clone or generate your own voice.
The original performance is filmed wide, with both people side by side under the mic. 4:3 keeps both of you and the mic in the frame; TikTok and Reels show it with bars above and below.
One adult per photo, face clearly visible, front-facing and well lit, ideally waist-up or full-length. Photos are checked before anything is charged: images with more than one face, cartoons, minors or nudity are turned away.
Only with their permission, and both people must be 18 or older. Photos of celebrities or public figures are not allowed.
They are sent to our AI providers only to check them and make your video. AIFlowMusic does not store them, and the video service deletes its temporary copy within 24 hours.
Pick who takes the left spot, choose 12 or 15 seconds, and let the booth do the rest.
Make your Hotel Lobby AI video