Quick answer
Upload two photos to Starrd's Hotel Lobby template and it drops both subjects into the orange COLORS booth — mic hanging between them, trading verses over 'Hotel Lobby' by Unc & Phew (Quavo and Takeoff). Both characters perform independently, so it reads like an actual back-and-forth rather than two people doing the same move. It costs 100 credits (about $5), takes two photos instead of one, and comes out vertical 9:16 at 15 seconds.
The same dance, any subject — tap one to make it with your photo.
Two photos in, and out comes both of them in the orange booth, trading bars. Mic hanging between them, no cuts, no camera moves, that flat COLORS wall behind. Here's exactly what you get:
Hotel Lobby
Two photos, one orange booth. You and whoever, trading bars over 'Hotel Lobby' — both subjects animated, mic between them, no prompt to write.
What the Sound Is
"Hotel Lobby" by Unc & Phew — the duo name Quavo and Takeoff put out together. Released 20 May 2022, produced by Murda Beatz, Keanu Beats and Fabio Aguilar. The COLORS performance is the version that travels: two of them in the booth, one mic, that orange wall.
"Unc and phew" is the uncle-and-nephew framing that became the whole identity, and it's what most people actually type into search — more than the song title, and far more than either name alone.
Is This Still a Trend?
Yes, and it's a strange one: it doesn't decay, it resurfaces.
COLORS posted the performance in June 2022. Then reposted it April 2025. Then October 2025. Then again in June 2026. Each repost catches a new wave, which is why the sound has over 428,000 videos on TikTok years after release rather than a spike-and-die curve.
This is the rare trend where being "late" doesn't matter. The COLORS booth is a format, not a moment — the orange wall reads instantly whether you post today or in six months.
The Fastest Way
Two photos, one tap.
Hotel Lobby
Drop two photos and both subjects land in the COLORS booth, trading verses over the track. 15 seconds, vertical, 100 credits.
What comes back: both subjects standing full-body in a seamless orange cyclorama, condenser mic hanging in the gap between them, each one performing independently — one leaning into the mic while the other bobs and waits. 15 seconds, vertical 9:16, the real track underneath.
Why Two Photos, Not One
Every other dance template on Starrd takes a single photo and animates a single subject. This one is the exception, and the reason is the format itself: a COLORS duo performance is the back-and-forth. One person in an orange booth is a different video.
So the template asks for two photos, reads each subject separately, and stages them side by side with a gap between them — then animates both. Not the same motion mirrored onto two bodies: genuinely independent performances, which is what makes it read as trading verses rather than a synchronised dance.
The casting is where the clips get good. A dog and a baby. You and your cousin. You and your dog. The pairing that has no business being in a rap booth is the one people send to their group chat.
Build It Yourself
If you'd rather assemble it by hand instead of using the template, here's the shape of it.
1. Get your two subjects into one frame. This is the step that breaks most attempts. You need a single still with both subjects standing full-body, side by side, clearly separated, against a flat orange background with a mic hanging between them. Feed both photos into an image model together:
A photorealistic VERTICAL 9:16 studio photograph of the TWO subjects from the reference photos performing together in a music session booth.LEFT subject: the first subject, standing TALL and FULLY UPRIGHT — body vertical, balanced, NOT sitting, NOT crouching. If the subject is an animal it stands on BOTH hind legs in a confident human-like posture. Arms or front paws raised and clearly separated from the body at about chest height, as if gesturing while rapping.RIGHT subject: the second subject, standing TALL and FULLY UPRIGHT on both feet, arms down and clearly separated from the torso, relaxed and confident.PRESERVE the exact identity of each subject — face, hair, skin tone, breed, fur, markings and clothing — exactly as in their reference photo. Only the pose and the setting change. Do NOT blend the two subjects together.They stand side by side with a clear gap of empty floor between them, neither overlapping nor touching, each forming ONE clean unobstructed silhouette. BOTH are shown FULL BODY from head to feet, feet flat on the floor, with headroom above and floor visible below. The pair is centered and fills MOST OF THE FRAME HEIGHT.SETTING: a seamless matte ORANGE studio cyclorama filling the entire background edge to edge — one flat continuous orange wall and floor, no corners, no furniture, no props. A single black condenser microphone hangs on a thin cable from the top of the frame in the gap between the two subjects. Soft even frontal studio lighting with warm orange bounce.Photorealistic, sharp focus, natural colors. Vertical 9:16. No text, no logos, no watermarks, no other people or animals.
2. Drive it with the performance. Take that still into a video model that accepts both an image reference and a video reference, and give it the booth clip to copy. Ask explicitly for full body, feet on the floor, a static locked-off camera and one continuous shot with no cuts.
3. Pick the right model. This is where most builds fall over. Dedicated motion-control tools are built to puppeteer one subject — hand them a two-person reference and they either reject it or lock onto whichever figure fills more of the frame and ignore the other. For a duo you want a video-to-video model that handles multiple characters natively. We use Seedance 2.0 for exactly this reason.
If your output has one subject performing and the other standing frozen like furniture, that's the single-subject limitation showing. It isn't a prompt problem and no amount of rewording fixes it — it's the wrong model for a duo.
4. Put the real track on afterward. Don't ask the video model to generate the audio. Render silent and lay the actual song over the top — a synthesised approximation of a real record always sounds worse than the record.
Common Mistakes
One group photo instead of two separate ones. The template needs a clean read of each subject before it can stage them. A group shot gives it one crowded read and the identities blur together.
Cropped or waist-up source photos. Both subjects need to be stood up full-body in the booth, so the more of each body the model can see, the better the staging.
Expecting them to move in sync. They shouldn't. The booth performance is a back-and-forth — one leads, the other waits. If both are doing identical moves, it stops looking like a duo.
Chasing the trend calendar. Don't. This one resurfaces every few months; the orange booth reads regardless.
FAQ
What is the Hotel Lobby COLORS trend? Quavo and Takeoff performing "Hotel Lobby" on A COLORS SHOW — two of them in COLORS' flat orange booth with one hanging mic, trading bars into the lens. Over 428,000 TikToks use the sound.
What song is it? "Hotel Lobby" by Unc & Phew, released 20 May 2022, produced by Murda Beatz, Keanu Beats and Fabio Aguilar.
Why two photos? Because the format is a duo. One person in an orange booth is a different video.
Can I pair a pet with a person? Yes, and it's usually the funniest version. Animals get stood upright automatically.
How much? 100 credits, about $5. Credits never expire.
