AI Hotel Lobby Trend: Prompt + How to Make It (2026)

The exact AI Hotel Lobby prompts we tested, the photo rules, the credits and the fixes. Copy the prompt, or make your orange booth video in one tap.

trendsai-videohow-totemplates

The AI Hotel Lobby trend is a short clip of two people or two pets in a flat orange booth, taking turns rapping into one hanging microphone. To make it with prompts, you need two full-body photos, a photo prompt that places them left and right in the booth, and a video prompt that animates the still with an original beat.

Below are the exact two prompts that made our example, the photo rules that decide whether it works, what it costs, and the fixes for the four things that usually go wrong. If you do not want to touch a prompt, the AI Hotel Lobby template runs both of them for you after you add two photos. Its Original song versions skip the prompts and put your two photos, or your two pets, into a real clip with the song.

Hotel Lobby example still (AI-made people)
Trend template

Hotel Lobby AI: Make the Hotel Lobby Trend Video

You and a friend, or your two pets, trade a rap into one hanging mic in a bright orange booth. Step into the real clip or the viral cat version with the original song, or get a new 10-second version with an original beat. No prompt to write. From $39.99, no subscription.

Where did the AI Hotel Lobby trend come from?

The set is real. Quavo and Takeoff performed "Hotel Lobby" on the YouTube show A COLORS SHOW, released on 17 June 2022: a flat orange wall and floor, one microphone on a cable, two performers side by side. By late September 2026 that video had more than 121 million YouTube views.

The AI version started in September 2026. XXL traced it to a TikTok clip of two cats in the booth in early September. Know Your Meme names a swap posted on 16 September as the first big one; it reached 1.7 million views in 12 days. A cat version on X passed 7.7 million views and 73,800 likes by 29 September, and Quavo reposted the original performance on 23 September. US Google searches for "ai hotel lobby" peaked on 25 September, according to the Google Trends data we pulled on 29 September.

DateWhat happened
17 June 2022The original performance goes up on A COLORS SHOW
Early September 2026A two-cat AI clip starts the format on TikTok (XXL)
16 September 2026First big AI swap on TikTok: 1.7M views in 12 days (Know Your Meme)
23 September 2026Quavo reposts the performance; music press covers the trend
25 September 2026US Google searches for "ai hotel lobby" peak

Takeoff died in 2022, and some fans asked people to keep the swaps respectful. The format is a joke about who is in the booth, not about him. Keep it that way.

Why does the Hotel Lobby prompt need two steps?

Because a video model is bad at inventing two specific people and good at animating a picture it is given. If you hand a video model two separate photos and one long prompt, it has to build the booth, place two faces, keep left and right, and invent motion, all at once. It usually drops one of those.

So the work is split:

  1. The photo step builds one still: both people, full body, in the orange booth, with the microphone between them. This is where faces, clothes and positions get fixed. You check this still before you spend anything on video.
  2. The video step animates that still. It only has to add motion and sound, and it keeps what the still shows.

On MakeViral the photo step runs on Nano Banana 2 and the video step on Kling 3.0. Kling tied for the best result in our 29 September test of this trend and had the cleanest lip movement for two voices.

What is the exact Hotel Lobby AI prompt?

These are the prompts, word for word, that made the example on the template page. Every line is there for a reason, and the reasons are in the next section.

Step 1: photo prompt (two people). Upload the left person first, then the right person.

Use the two uploaded images as the base. The person in the first image stands on the LEFT. The person in the second image stands on the RIGHT. Keep each person's face, facial features, skin tone, hair, body shape and clothing exactly the same as in their own photo. Do not blend the two people and do not swap their clothes. Show both people full body, head to feet, standing upright side by side with a clear gap of floor between them. They stand inside a small seamless studio booth with a flat matte orange wall and floor, no corners and no furniture. One black studio microphone hangs on a thin cable from the top of the frame, in the gap between them. The left person has one hand raised as if about to rap. The right person stands relaxed with arms loose. Soft, even frontal light with a warm orange bounce. Photorealistic, sharp, vertical 9:16. No text, no logos, no watermarks, no other people.

Step 2: video prompt (10 seconds, with sound). Use the still from step 1 as the first frame.

Two friends perform a playful hip-hop duet in a small orange studio booth with one hanging microphone between them. Keep both faces, outfits and positions exactly as in the image. The left person always stays on the left and the right person always stays on the right. The left person leans toward the microphone and raps first with loose hand gestures while the right person nods to the beat and reacts. Then the right person takes the lead while the left person hypes them up. The timing feels natural and conversational, never mirrored. One locked-off, centered, wide shot. No cuts, no zoom, no camera movement. Audio: an original upbeat hip-hop beat with both voices rapping short, clean, playful lines. No text, no logos, no extra people.

The pet version of the prompt

The trend started with cats, and pet pairs still get some of the biggest numbers. The photo prompt stands both animals upright:

Use the two uploaded images as the base. The animal in the first image stands on the LEFT. The animal in the second image stands on the RIGHT. Keep each animal's breed, face, eye color, fur color, markings, collar and body shape exactly the same as in its own photo. Do not blend the two animals. Both animals stand upright on their hind legs in a confident, human-like pose, front paws loose like arms, with a clear gap between them. Show both full body, ears to paws. They stand inside a small seamless studio booth with a flat matte orange wall and floor. One black studio microphone hangs on a thin cable from the top of the frame between them. Soft, even frontal light with a warm orange bounce. Photorealistic, vertical 9:16. No text, no logos, no people.

And the video prompt gives them small cartoon voices:

Two pets stand upright in a small orange studio booth with one hanging microphone between them. Keep each animal's face, fur and markings exactly as in the image, and keep their positions. The left pet leans to the microphone first and "raps" with small paw gestures and head bobs while the right pet watches and reacts. Then the right pet takes its turn while the left pet bobs to the beat. Independent, conversational movement, never a synchronized dance. Static, centered, full-body shot. No cuts. Audio: an original funny hip-hop beat with small cartoonish pet voices. No extra paws, no merged bodies, no text, no logos.

Which parts of the prompt matter most?

Four phrases do most of the work. If you edit the prompts, keep these.

PhraseWhat it prevents
"The person in the first image stands on the LEFT"Swapped positions. The model keeps upload order only when you say it.
"Do not blend the two people"One face that is half of each person. This is the most common failure.
"flat matte orange wall and floor, no corners and no furniture"A normal room painted orange. The trend reads as the booth only when the wall is seamless.
"never mirrored" and "conversational"Both people rapping the same move at the same time, which looks like a glitch.

"One locked-off, centered, wide shot. No cuts" matters too. The original performance is one static camera, and a moving camera makes the clip look like a different trend.

"An original upbeat hip-hop beat" is deliberate. The viral clips reuse the real song; a generated copy of it would put copyrighted audio in your file. Ask for original music and add the trending sound in the app when you post, if you want it. If you want the real song in the file, use the Original song version of the template instead: it starts from the real clip, so nothing is generated in its place.

How do you make the Hotel Lobby video step by step?

This is the one-tap route on MakeViral. The screenshots are from the live site on 30 September 2026, taken on a 390-pixel phone screen.

1. Add the two photos. Open the Hotel Lobby template and press "Make this video". Pick "Two people" or "Two pets" for the two-step route in this guide, then add the left photo in slot 1 and the right photo in slot 2. You can take them with the camera or pick them from your library. (The page opens on "Original song", which works differently: see the end of this section.)

MakeViral Hotel Lobby create page with two labeled photo slots, left person and right person
Slot 1 is the left side of the booth, slot 2 the right. One person per photo.

2. Make it yours (optional). Open "Make it yours" to pick a song style, write the lines they rap, set a mood or a language, and choose matching outfits. It does not change the price. Short lines work best, because the two people take turns line by line.

Make it yours panel with song style chips, a lyrics box with example lines and mood options
Song style, lyrics and mood. The Generate button shows the price before you spend anything.

3. Press Generate. The button shows the credits first. MakeViral builds the booth still with your photos, then Kling 3.0 animates it for 10 seconds with sound. You can close the page; the video lands in My videos, and a notification tells you when it is ready if you allowed them.

4. Download or share it. The result is a 9:16 MP4 with sound. "Share to TikTok" opens your phone's share sheet with the video file and copies a caption for you to paste. "Download" saves the file.

Finished Hotel Lobby video with two AI-made people in the orange booth, and Share to TikTok, Download and Share buttons
A finished run from our test account. The two people are AI-made, not real persons.

If you would rather edit the prompts yourself, the template page has "Edit the prompts yourself": it opens AI Photo with the photo prompt filled in, and each finished still has a button that opens AI Video with the video prompt ready.

The Original song versions

These two versions make no still and need no prompt. You add two photos, left first, and top-quality AI video models put them into a real clip:

  • Original song uses the real 2022 performance. Photo 1 replaces the performer on the left, in the striped shirt, and photo 2 the one on the right. The moves, the camera cuts, the lip sync and the song stay.
  • Original song, pets uses the most viral AI version of the trend: two AI-made cats by @therealmeowfia on TikTok, posted on 23 September 2026 and at 6 million views by 12 October. Pet 1 replaces the black cat on the left, who raps, and pet 2 the ginger cat on the right, who dances. The reel's song goes back on after the run, and you can turn it off on the result screen.

Both make a 15-second clip, wide 16:9 like the clip they copy, and both cost more than the two-step route (see the prices below).

Characters, toys and cartoons: the whole-body swap prompt

The Original song version also takes a toy, a cartoon or game character, a figure or an object, not only a person. The first version of our prompt swapped only the head of a character: the green cactus figure in our test kept the performer's orange shirt and human arms. Two changes fixed it, and this is the prompt that passed, word for word (Seedance 2.5, with the real clip as [Video 1]):

  • Name each performer by place, not by clothes. "The man on the right in the orange knit shirt" told the model the shirt mattered, so it stayed. "The man on the right" does not.
  • Ask for the whole body. Its own head, torso, arms, paws, legs, feet or a base replace all of the performer, with no human arms, hands or shirt unless the photo shows them. A subject with no limbs sways and bounces with its whole body.
Recreate the reference video [Video 1] shot for shot: the same orange studio booth, the same hanging microphone, the same light, the same camera angles, cuts, timing and moves. The ONLY change: [Image 1] replaces the man on the left, and [Image 2] replaces the man on the right. Each photo can show a person, a pet, a toy, a cartoon or game character, a figure or an object: whoever or whatever it shows takes that man's place with its WHOLE body, not only the head.

Use [Video 1] only for the place, the camera, the cuts, the timing and the moves. Do not copy the two men from [Video 1]: nothing of them stays in any shot, not their faces, skin, hair, hands, shirts, jeans, shorts, glasses or jewelry.

The one on the left = [Image 1], from head to feet in EVERY shot. If [Image 1] shows a person: copy the exact face - face shape, eye color and eye shape, eyebrows, nose, mouth, skin tone, age, hair color, hair length and hairstyle, and any beard, moustache, glasses or freckles exactly as in the photo - and the clothes from [Image 1], with no white sunglasses unless [Image 1] shows them. If it shows anything else: that exact subject, with its own shape, proportions, colors, surface, material and art style; its own whole body, whatever it has (head, torso, arms, paws, legs, feet or a base), replaces ALL of the man's body, with no shirt, no shorts, no jeans and no human arms, hands or skin unless [Image 1] shows them; it raps and points with its own limbs, or sways and bounces with its whole body when it has none; a blocky figure stays blocky, a plush stays soft, a plant stays a plant.

The one on the right = [Image 2], from head to feet in EVERY shot, copied from [Image 2] in the same way: a person in the clothes from [Image 2] with no tinted glasses unless [Image 2] shows them, anything else with its own whole body and no human arms, hands, shirt or shorts unless [Image 2] shows them.

When a photo shows only a head or an upper body, add the rest so it matches. Both copy the rapping, the gestures and the hand moves of the two men in the way their own bodies allow, and a mouth moves with the song when they have one. The same two in every frame, from the first shot to the last, so their friends instantly recognize them. Never morph, never drift back to the original men. Do not mix them: [Image 1] is only the one on the left, [Image 2] is only the one on the right.

Shots, matching the reference:
1. Wide: both stand in the orange booth by the hanging microphone.
2. Medium two-shot: they rap in turn into the microphone, with loose hand moves, and react to each other.

Anatomy rules: one head each, on its own body. A face never appears on hands or clothes. Exactly two performers. No text, no logos.

The same rules work for any clip. On the AI video swap page, "Me" and "2 people" take a toy or a character photo the same way.

What photos work best for the AI Hotel Lobby trend?

The photo decides the result more than the prompt does.

  • Full body, head to feet. The booth shot is full length. A cropped selfie leaves the model to invent legs and shoes, and it often gets them wrong.
  • One person per photo. Two photos of one person each keep the faces apart.
  • Plain clothes with one strong color. They read well against orange. An orange top disappears into the wall.
  • Good, even light on the face. Harsh shadows turn into odd features in motion.
  • For pets, the whole animal in focus. The prompt stands it upright, so the model needs to see the legs and tail.

Use photos of yourself, of adults who agreed to be in the video, or of your own pets. Never a celebrity or a stranger.

How much does an AI Hotel Lobby video cost?

On 12 October 2026 the prices on MakeViral were:

RouteSettingsCredits
One-tap template, Two people or Two pets, Kling 3.010 s + 1 scene photo, charged once56
One-tap template, Original song or Original song, pets, H3 Max Recast15 s on a real clip, with the song180
By hand in AI Video, Seedance 2.0 Fast10 s, 720p + 1 photo68
By hand in AI Video, Veo 3.1 Fast8 s, 720p + 1 photo14

At the Starter plan rate (700 credits for 49.99 dollars a month), 56 credits is about 4 dollars and 180 credits about 13 dollars. Veo 3.1 Fast is the cheapest route, but it makes 8 seconds instead of 10, and you run the two steps yourself. A failed video gives its credits back. Current prices are always on the template page and on the Generate button.

What goes wrong, and how do you fix it?

ProblemFix
One face looks like a blend of both peopleUse sharper photos with one person each, then make the still again.
Left and right are swappedRemove both photos and add them again, left person first. Check the still before you animate it.
The booth has corners or furnitureGenerate the still again. The prompt asks for a seamless wall and a second run usually gets it.
Both people rap at the same timeKeep "never mirrored" and "conversational". Shorter lyric lines help.
A hand warps near the microphoneCommon in 10-second clips. A second run, or 8 seconds instead of 10, often fixes it.
The video model refuses the photoSome models refuse some real faces. The credits come back; try Veo 3.1 Fast.

How do you post the Hotel Lobby video so it travels?

The format already does the hard part: viewers know the trend in the first frame, so the only question is who is in the booth. Put that answer in the caption or in on-screen text: "me and my sister", "our two cats", "the night shift at work".

  • Post both people's accounts. A duet gets reposted twice when both people tag each other.
  • Contrast sells. Most big copies in September 2026 paired opposites: a tiny cat and a huge one, a calm person and a hyper one.
  • Add the trending sound in the app if you made Two people or Two pets and want the original song, from TikTok's or Instagram's music library. Your file keeps its original beat either way. With the Original song versions, the song is already in the file.
  • Move fast. Visual trends like this one peak within two to four weeks. For the bigger picture on hooks, timing and trends, read how to go viral on TikTok.

For pet accounts and duo pages that want a steady stream of ideas, the faceless TikTok ideas list pairs well with the month's AI video templates, such as the AI dancing dog video and the 80s AI trend.

Hotel Lobby example still (AI-made people)
Trend template

Hotel Lobby AI: Make the Hotel Lobby Trend Video

You and a friend, or your two pets, trade a rap into one hanging mic in a bright orange booth. Step into the real clip or the viral cat version with the original song, or get a new 10-second version with an original beat. No prompt to write. From $39.99, no subscription.

FAQ

Questions people ask

What is the AI Hotel Lobby trend?
It is a viral AI video format from September 2026. Two people or two pets stand in a flat orange studio booth and take turns rapping into one microphone that hangs between them. It copies the set of Quavo and Takeoff's 2022 "Hotel Lobby" performance on A COLORS SHOW, with new performers swapped in.
What is the best prompt for the AI Hotel Lobby video?
Use two prompts, not one. A photo prompt first puts both people into the orange booth and fixes who stands left and right. A video prompt then animates that still for about 10 seconds with an original beat. Both prompts are in full in this guide, exactly as they made our example clip.
Can I make the Hotel Lobby video with one photo of two people?
It works better with two photos, one person each, full body. A single group photo gives the model less detail per face, and it often blends the two people or cuts off the legs. Use the left person's photo first, then the right person's.
How much does it cost to make the Hotel Lobby AI video on MakeViral?
On 12 October 2026 the Two people and Two pets versions of the one-tap template cost 56 credits: the scene photo plus a 10-second Kling 3.0 video with sound, charged once. The Original song and Original song, pets versions cost 180 credits for 15 seconds on a real clip with its song. At the Starter plan rate of 700 credits for 49.99 dollars, that is about 4 dollars and about 13 dollars. A failed video gives its credits back.
Can I use the real Hotel Lobby song?
Yes, with the Original song version of the template. It puts your two photos into the real 2022 clip, so the song and the lip sync in your file are the real ones. Original song, pets does the same for two pets with the most viral AI cat version of the trend, with that reel's song. The two-step prompts in this guide ask for an original beat instead, so their lip movement follows the generated lines, not the song. The song belongs to its owners, and MakeViral gives you no rights to it.
Can I put a toy, a cartoon or a game character in the Hotel Lobby video?
Yes. Since 30 October 2026 the Original song version takes any subject: a person, a pet, a toy, a cartoon or game character, a figure or an object. The whole subject takes the performer's place, from head to feet, in its own shape, colors and style. In our test, two blocky game figures replaced both performers in every shot, with nothing of the two men left. The full prompt is in this guide.
Is it okay to put a celebrity in the booth?
No. A fake video of a real, famous person can break platform rules and that person's rights, and some video models refuse well-known faces. Use yourself, adults who agreed to it, or your own pets.
Alex Deckin
Editor, MakeViral

Alex Deckin is on the MakeViral team and is responsible for the guides on this blog. He is based in Austin, Texas, studied computer science, and has spent four years studying AI technologies, with a focus on generative AI and AI video generation.

Make the video, not just the plan

Pick a format, paste a prompt or a product URL, and download a 9:16 video that is ready to post.

No watermark. A failed video gives its credits back. Secure checkout by Stripe.