How to Make the AI Hotel Lobby Trend From Two Photos
The AI Hotel Lobby trend step by step: which two photos to use and in which order, what happens in the 14 seconds, your clothes or the clip's, the sound and the price.
A flat orange wall, one silver microphone hanging from the ceiling and two people trading a rap. Since September 2026 that booth from Quavo and Takeoff's COLORS performance of "Hotel Lobby" has been filled with friends, couples, coworkers and cartoon characters. Here is how to put you and a friend in it from two photos, without a prompt or an editing app.
What is the AI Hotel Lobby trend?
COLORS released the "Hotel Lobby" performance in June 2022. In mid-September 2026 AI swaps of the booth took off on TikTok: the first big one put two characters from Suits at the mic, and within days everyone was making their own. The Hotel Lobby template follows the camera, timing and moves of a viral swap of the booth and puts your two people in it, with the trend sound.
How to make it
- Open the Hotel Lobby template.
- Add photo 1: the person on the left, who stands calm and takes over the verse.
- Add photo 2: the person on the right, who raps at the mic first and turns their back.
- Pick a version and generate. Every version renders the whole 14-second clip.
- Download and post. The MP4 already has the sound, lined up from the first second.
What happens in the video
- 0–2.5 s: a locked-off 4:3 shot; the right performer raps at the mic with an arm out at the camera, the left one stands with hands clasped.
- 2.5–5.5 s: the right performer counts the beat on their fingers; the left one rubs their hands and smiles.
- 5.5–8.5 s: the right performer leans in, rolls both hands outward and turns away.
- 8.5–11.5 s: back to the camera, an arm over the head, while the left performer takes over.
- 11.5–14 s: face to face across the mic for the last lines.
Your clothes or the clip's?
By default you both perform in the clothes from your photos. Under Advanced, "Outfit from the video" dresses you like the clip instead: the long white coat on the left, the orange hat and bare chest on the right. Same price either way, so it's worth trying both.
Photos that work best
- Waist-up and facing forward. You are framed from the thighs up and turn your heads, so the whole face should be visible.
- No orange tops. An orange shirt disappears into the wall; one strong other colour reads best.
- One person per photo, sharp, in good light, no sunglasses.
- Pick your side. Swap the photos before you generate if you want the other part.
- Ask first. Use your own photo and one of someone who said yes; never a stranger or a public figure.
What does it cost?
Balanced is 10 credits at 480p and 25 at 720p. Premium is 25 credits at 480p, 55 at 720p and 95 at 1080p, and your account's first full 480p video costs just 20 credits. Without an account, one video starts at $7.99. No watermark, and failed renders are refunded automatically.
FAQ
Is this the original COLORS video? No. It is a new AI take of the booth with your two people; we don't use or host the COLORS recording. Can I use one photo of us together? The template takes two photos, one per person; crop a photo of you two into two photos if that's all you have. Do I need to add the song? No, the sound is already in the MP4.