The same set and shots
The teal wall, the hanging microphone and the low platform stay, with the same two-shots, close-ups and wide shots in the same order.
For the 58th Floor COLORS trend
58th Floor AI Video Generator
Put two people on the teal set of the performance everyone is remaking. One raps first, then hands over, and the shots cut just like the original, around one hanging microphone.
- 1–2 photos
- 15 or 30 seconds
- 720p · 9:16
- No watermark
01Add one or two photosAdults · one person per photo · one photo is enough
Left personRaps first, by the mic
Right personTakes over the verse
A clear, front-facing photo works best. One adult per photo; the clothes come from the photo too.
02Pick the length
03Pick the sound
900 credits · 15 s · 720p · 9:16
Sign in to add photos. Credits stay valid for 6 months. Get creditsFrom two photos to the verse
Two photos in, 15 seconds out. There is no 30-second example yet: that version adds 15 seconds before this part.
15 seconds · Two photos · Original track
Left person
Right person
The example uses two AI-generated people, not real ones.
Why it looks like the performance
The set, the microphone and the order of the shots are fixed. Your photos decide who is on stage.
The hand-over stays
The left person raps first and the right person takes over, while the other keeps quiet and moves to the beat.
One photo is enough
Add just one photo and you rap with yourself, like twins. Add two for a duo.
Credits back if it fails
A render that fails returns to your balance. No ticket, no waiting.
How to make a 58th Floor AI video
Three steps, usually 10 to 30 minutes.
- 1Add one or two photos
One adult per photo, facing the camera, with the face clearly visible. A full-body photo also gives the wide shot the right clothes.
- 2Pick the length and the sound
15 seconds or 30 seconds. Keep the original track, or choose Silent and add your own music when you post.
- 3Render, download, post
720p, 9:16, no watermark. Post it to Reels, TikTok or Shorts.
What is the 58th Floor AI trend?
The 58th Floor trend comes from "58th Floor", a song by Anderson .Paak and Cordae, and from the way they performed it on COLORS: one teal wall, one hanging microphone and two friends trading verses.
People remake it with AI by putting themselves, a friend or a partner on that set, usually with the 15 seconds where one rapper hands over to the other. This page does the same from your photos and keeps the set, the shots and the track.
Where did the 58th Floor trend start?
"58th Floor" is on Heavy Is the Crown, the album Anderson .Paak and Cordae released together on October 23, 2026. A week later COLORS, the music platform known for its plain single-colour sets, released their performance of it as A COLORS SHOW.
COLORS framed the performance around friendship, with a post about the people who bring out your best and your worst. Within days AI template apps offered it as a 15-second clip and a longer 30-second cut.
Timeline
- October 23, 2026Anderson .Paak and Cordae release Heavy Is the Crown, with "58th Floor" on it.
- October 30COLORS releases "58th Floor" as A COLORS SHOW, on a teal set with one hanging microphone.
- Early NovemberAI template apps add 15-second and 30-second versions, and people start remaking it with their own faces.
Sources: The release on Apple Music · The performance on 24Hip-Hop · The COLORS post on Threads · The template on Starrd
What happens in the 15 seconds
The rough timing of the 15-second version; a cut can land about a second early or late. The 30-second version puts 15 more seconds before it: the left person alone, a close-up of them, then a wide shot of both.
| 0–4 s | Both from the waist up: the left person raps by the mic, the right person listens and moves to the beat |
|---|---|
| 4–5.5 s | They turn to face each other, and the right person takes over at about 5 seconds |
| 5.5–8.7 s | Close-up of the right person rapping, with the mic at the edge of the frame |
| 8.7–11.5 s | Wide shot, head to toe on the low platform: the right person raps and the left person dances |
| 11.5–15 s | Back to both from the waist up: the right person raps at the left person, who nods along |
The photos that work best
Use one adult per photo, facing the camera, with the whole face sharp, evenly lit and uncovered. The close-ups show the faces large, so a clear face matters for both people.
The clothes come from your photos, and the wide shot shows everyone head to toe, so a full-body photo works best. Anything the photo does not show is continued in the same everyday style. Keep one person in each photo: a group photo does not count as two.
Pricing - AI Video Maker Credit Packs & Plans
Pick the plan that matches your pace
Subscribers get +20% extra credits at the same price
900 Credits
$9.9/ one-time
≈ 18 videos5s · 480p each
Subscribers receive: 1,080
- Up to 450 fast images
- Up to 18 basic videos
- Every image and video model
- Valid for 6 months
- No subscription, never auto-renews
1,900 Credits
$19.9/ one-time
≈ 38 videos5s · 480p each
Subscribers receive: 2,280
- Up to 950 fast images
- Up to 38 basic videos
- Every image and video model
- Valid for 6 months
- No subscription, never auto-renews
5,500 Credits
$49.9/ one-time
≈ 110 videos5s · 480p each
Subscribers receive: 6,600
- Up to 2,750 fast images
- Up to 110 basic videos
- Every image and video model
- Valid for 6 months
- No subscription, never auto-renews
12,000 Credits
$99.9/ one-time
≈ 240 videos5s · 480p each
Subscribers receive: 14,400
- Up to 6,000 fast images
- Up to 240 basic videos
- Every image and video model
- Valid for 6 months
- No subscription, never auto-renews
Subscriptions renew automatically at the price shown until you cancel, and we'll tell you before any price change. Cancel anytime in Settings → Billing; you keep access until the paid period ends. Credit packs are one-time and expire 180 days after grant. Prices include any tax collected at checkout. Refunds & Billing · Terms · Privacy Policy · Contact Us · Pricing
aiMakeVideo is an independent AI video platform and is not affiliated with the model providers.
Questions
58th Floor AI FAQ
What is a 58th Floor AI video?
It is a vertical video of two people trading verses at one hanging microphone on a teal studio set, starring the people in your photos. 15 seconds costs a flat 900 credits and 30 seconds costs 1,900.
What song is in the 58th Floor trend?
"58th Floor" by Anderson .Paak and Cordae, from their album Heavy Is the Crown. The clip's track comes from their COLORS performance. Your video comes with that track unless you choose Silent.
Who is in the original video?
Anderson .Paak and Cordae, performing on COLORS. Their faces and clothes are not used here: we rebuild the set and replace both of them with the people in your photos.
What is COLORS?
COLORS, also written COLORSxSTUDIOS, is a music platform that films artists performing on plain single-colour sets. This generator is not made by or connected with COLORS.
Will the lips match the words?
Roughly, not word for word. Each person's mouth moves on their turn and rests while the other raps, but the syllables will not line up exactly, especially on a fast verse. A sharp, front-facing photo helps.
Does the track have explicit lyrics?
Yes. The song has explicit language, including in the 15 seconds this page uses. If that is a problem for where you post, choose Silent and add your own music.
Can I put my pet in it?
Not yet. This generator works from photos of adults. To put your pet into a dance video instead, try Dime Dime Baby.
How does this generator work?
Add one or two photos, then pick the length and the sound. We replace the two people in a fixed version of the clip: the set, the microphone and the order of the shots stay as they are, and the faces, hair and clothes come from your photos.
What photos work best?
One adult per photo, facing the camera, with the face sharp and evenly lit. A full-body photo also gives the wide shot the right clothes. Avoid sunglasses, masks, heavy filters and group photos.
What happens if I add only one photo?
You play both people, like twins, in the same clothes, in both the 15-second and the 30-second version. Add two photos to put two real people in the clip.
Which side should I pick?
The left person raps first and has the mic beside them. The right person takes over at about 5 seconds and has the close-up in the 15-second version. In the 30-second version the left person also gets a solo and a close-up before that.
How much does it cost?
A flat 900 credits for 15 seconds and 1,900 credits for 30 seconds, with one photo or two, with sound or without. The smallest pack is $9.90 and covers a 15-second video, and a failed render returns to your balance.
Can I change the place or the clothes?
The teal set stays: it is what makes the clip recognisable. The clothes come from your photos, so wear what you want to see in the video.
Is the sound included?
Yes. By default the original track runs under the whole video. Choose Silent if you want to add your own music when you post.
How long does it take, and what if it fails?
Usually 10 to 30 minutes, and up to an hour when the queue is busy. You can close the tab, and the finished video waits in My videos. A failed render returns the credits straight to your balance.
Are you affiliated with the song, the artists or COLORS?
No. This is an independent fan tool with no connection to Anderson .Paak, Cordae, the rights holders, COLORS, TikTok, Instagram or any creator.
More AI video tools
Every tool turns your photos into a short vertical video. One credit balance works on all of them.
An independent fan tool. Not affiliated with Anderson .Paak, Cordae, the song's rights holders, COLORS, TikTok, Instagram or any creator.
Ready for your verse?
One or two photos. One microphone. Fifteen or thirty seconds.
Make my 58th Floor video