58th Floor AI Video

58th Floor AI prompt

Skip the prompt: make it from two photos

Add a photo for the hook and one for the verse; the teal studio, the hanging mic and the camera cuts come from the trend's clip.

Who you play

Add both photos

Example · Made with this template · 15 s · 9:16

See the first 2 seconds free. The full video is 900 credits.

One person per photo, face clearly visible. JPG, PNG or WebP up to 10 MB.

  • No app
  • No watermark
  • Credits back if a video fails
See pricing

Copy the 58th Floor AI prompt below into an AI video tool that takes two reference photos. We ran it once in Wan 3.0 with two portraits, so you can see what it got right, what it missed and what to change next. Rather not write a prompt? Make it here from two photos; the teal studio, the mic and every camera cut come from the trend's own clip.

  • For: AI video tools that take two reference photos
  • Length: written for a 10-second clip
  • Shape: set the tool to 9:16
Copy and paste

The full 58th Floor AI prompt

Paste this into an AI video tool that takes reference photos. Add the photo of whoever opens with the hook first and the verse rapper's photo second, and set the shape to 9:16. If your tool calls the photos something other than Image1 and Image2, change those two words. In our run the model followed the shots in order but left the mic high above them in the wide shot (see below).

For a tool that takes only one photo, use a photo of the two of you together and change the second sentence to "the person on the left raps first and the person on the right takes the verse". We have not run that version.

Prompt

Vertical 9:16 music performance video in the style of a COLORS session: a seamless teal studio with one vintage silver microphone hanging from the ceiling on a long black cable, soft even light. Image1 raps first and Image2 takes the verse; keep each person's face, hair and clothes from their image the whole time. 0-4 s: waist-up two-shot, Image1 raps the hook toward the camera with hand gestures while Image2 stands beside them, nodding along with mouth closed. 4-5 s: Image2 turns to face Image1. 5-8 s: close-up of Image2 rapping the verse into the hanging microphone. 8-10 s: wide full-body shot of both on a low teal platform, Image2 still rapping while Image1 dances and nods along, impressed. Exactly two people and one microphone, five fingers on each hand, no captions or on-screen text.

  1. The room — "Seamless teal studio", "one vintage silver microphone hanging from the ceiling" and "soft even light" describe the COLORS look without naming a real performance or using footage from it.
  2. The two roles — "Image1 raps first and Image2 takes the verse" says who does what, and "keep each person's face, hair and clothes" is there to stop the two of them drifting toward each other.
  3. The shot list — The four timed parts set the order of the camera: two-shot, the turn, the close-up at the mic, the wide shot. Without times, most models pick one framing and stay on it.
  4. The clean-up line — "Exactly two people and one microphone", five fingers and no captions cut down on a third person, a second mic, warped hands and stray text.
Shot by shot, then a real run

What the 58th Floor prompt asks for, and what Wan 3.0 made of it

Four shots in the prompt, one real run: here is which shots Wan 3.0 followed when we gave it two portrait photos.

The 10 seconds, shot by shot

If your tool makes shorter clips, keep the order and squeeze the timing. The 15-second clip on this site has the same four shots, with cuts at 5.5, 8.7 and 11.5 seconds.

  1. 0 s–4 s

    Waist-up two-shot: the hook rapper talks to the camera and gestures, the verse rapper nods along with mouth closed.

  2. 4 s–5 s

    The verse rapper turns to face the hook rapper.

  3. 5 s–8 s

    Close-up of the verse rapper at the hanging mic.

  4. 8 s–10 s

    Wide full-body shot of both on a low teal platform, the hook rapper dancing and nodding along.

Our run in Wan 3.0

Mostly right: both people kept their look, the teal room and the hanging silver mic were there, and the shots came in the order asked, with the verse rapper's close-up at the mic. The wide shot at the end is where it slipped; see Missed below.

Prompt we used

Vertical 9:16 music performance video in the style of a COLORS session: a seamless teal studio with one vintage silver microphone hanging from the ceiling on a long black cable, soft even light. Image1 raps first and Image2 takes the verse; keep each person's face, hair and clothes from their image the whole time. 0-4 s: waist-up two-shot, Image1 raps the hook toward the camera with hand gestures while Image2 stands beside them, nodding along with mouth closed. 4-5 s: Image2 turns to face Image1. 5-8 s: close-up of Image2 rapping the verse into the hanging microphone. 8-10 s: wide full-body shot of both on a low teal platform, Image2 still rapping while Image1 dances and nods along, impressed. Exactly two people and one microphone, five fingers on each hand, no captions or on-screen text.

Settings
Reference photos to video, 480P, 9:16, 10 seconds, two portraits, first result
Wait
About 3 minutes
Came out right
Faces, hair and tops for both; the teal room and the hanging silver mic; the turn at about 4 seconds; the close-up of the verse rapper at the mic; the wide shot on a round teal platform
Missed
In the wide shot the mic hangs far above their heads and the hook rapper only smiles instead of dancing; the close-up starts about a second late; trousers and shoes were made up, because the photos stop at the chest
A woman with a red bob in a mustard sweater talks to the camera beside a bearded man in a green jacket under a hanging mic, from the Wan 3.0 test run
The two portraits used for the test: a woman with a red bob and round glasses in a mustard sweater, and a bald bearded man in a green jacket over a white tee
Photos we used

Fixing common problems with the 58th Floor prompt

These are starting points to try. Change one thing per run, so you know what helped.

  • In the wide shot the mic floats far above them

    Try next
    Say where it hangs: "the microphone hangs at mouth height between them in every shot, including the wide shot"
    Leave alone
    The room line

  • The wrong person raps the verse

    Try next
    Check the photo order: the hook rapper first (Image1), the verse rapper second, and name the verse rapper again in the close-up line
    Leave alone
    The rest of the shot list

  • The camera never cuts

    Try next
    Start each timed part with the framing word ("two-shot", "close-up", "wide shot") and add "hard cut between shots"
    Leave alone
    The timings

  • The two faces start to look alike

    Try next
    Use photos taken in different light and add "two different people who do not look alike"
    Leave alone
    "Keep each person's face, hair and clothes"

  • Trousers or shoes you weren't wearing

    Try next
    Use full-body photos, or name the whole outfit in the prompt
    Leave alone
    The face and top, if they held

  • A second mic, a third person or text on screen

    Try next
    Move "exactly two people and one microphone" to the start of the prompt
    Leave alone
    The clean-up line at the end

The 58th Floor prompt vs two photos here

A prompt lets you describe any room and any performers, and the camera is the model's own idea. Photos here give you the trend's own clip and cuts, without writing anything.

This promptIn an AI video toolTwo photos hereNo promptUse my photos
What you writeThe prompt above, adjusted until it looks rightNothing
The camera cutsWhatever the model makes of the shot list; in our run, in order but lateThe clip's four shots, cut at 5.5, 8.7 and 11.5 seconds
The mic in the wide shotIn our run, far above their headsWhere it hangs in the clip
The songWhatever sound your tool adds, or noneThe song underneath, already in the video
ShapeWhatever you set; ours was 9:16 at 480P, 10 seconds15 seconds, 9:16, 720p

58th Floor AI prompt questions

See also

More 58th Floor guides

Skip the prompt

Two photos, the trend's own studio and cuts.

Make our 58th Floor video