How many photos do I need for Tiny Desk AI?

One is enough, for the lead singer. You can add up to four more, in any mix: the keyboard player, the two backing singers and the saxophonist.

What happens to the roles I leave empty?

Leave a role empty to keep that performer as they are, so you still get a band of five.

How long is the video and what shape is it?

15 seconds, vertical 9:16, as an MP4.

Does the video have sound?

Yes. It comes with the template's music, 15 seconds of the trend's track. Mouth movement follows the performance but will not match every word.

No. A filter changes video you filmed. We make a new clip in which the people in your photos play their roles in the band, and the roles you leave empty stay as they are.

No. This scene is built for people. Use an adult who agreed to be in the video.

Can I order a Tiny Desk AI video today?

Yes, orders are open. 480p is $9.99, 720p is $19.99 and 1080p is $39.99. You pay once per video. There is no subscription.

Why are the backing singers less sharp?

They stand by the shelves and are smaller in the frame, so their faces come out less clear than the lead's.

How to Make a Tiny Desk AI Video From Your Photos (2026)

Tiny Desk is a small-concert clip: a band squeezed behind an office desk, with books and plants all around. A Tiny Desk AI video puts the people in your photos into it. You get a 15-second vertical MP4 that is AI generated and comes with the template's music. You pay once per video, from $9.99, with no subscription.

Video de YouTube de Put Me In: Our Short: a finished Tiny Desk AI video, then the three steps on the site. — Our Short: a finished Tiny Desk AI video, then the three steps on the site.
Put Me InVer en YouTube

What You Get

An office desk crowded with books and plants, lit by a warm lamp.
A lead singer who gets the close-up and cracks up mid-song, the person from your first photo.
A keyboard player at a red keyboard, from your second photo if you add one.
Two backing singers by the shelves, from your third and fourth photos if you add them.
A saxophonist who plays the last few seconds, from your fifth photo if you add one.
Leave a role empty to keep that performer as they are.
A 15-second MP4 in 9:16, with the template's music and an AI generated label on the file.

What the Trend Is

Tiny Desk is a clip where a full band plays a song in a space that is far too small for it. It works as a meme because the performance feels big and the setting is an ordinary office. People like to picture themselves at the microphone, or to give each friend a role. Many people search for it as Outstanding Tiny Desk, after the song in the original clip, and it is the same trend.

The AI version rebuilds the scene with your photos. It is meant as a joke or a gift for friends who like to play music, not as a realistic concert. Put Me In has no connection with the original performers, the song, the show or anyone who owns them. "Tiny Desk" only describes the trend. For the background, read what the Tiny Desk AI trend is.

What to Prepare

One photo is required, and you can add up to four more, in any mix.

Photo one, the lead singer. Required.
Photo two, the keyboard player. Optional.
Photo three, the left backing singer. Optional.
Photo four, the right backing singer. Optional.
Photo five, the saxophonist. Optional.
One adult per photo, face fully visible, front-facing, with even light.
No sunglasses, masks or heavy filters.
Yourself, or friends who agreed to be in it.

Photos of public figures or anyone under 18 are not allowed. More in how to pick photos.

If you are casting friends, decide the roles first. The lead singer gets the longest close-up, so give that spot to the clearest photo. The backing singers are the smallest in the frame.

Steps

Orders are open on the Tiny Desk page:

Open the Tiny Desk page and upload the lead singer's photo.
If you want, upload photos for the keyboard player, the backing singers and the saxophonist.
Pick a quality: 480p ($9.99), 720p ($19.99) or 1080p ($39.99).
We check the photos. If one does not pass, nothing is charged and you can try another.
Pay once. There is no subscription.
The clip appears under My videos in a few minutes.

Common Mistakes

A group photo for a role. The model may not know who you mean. Use a photo of one person for each slot.
Expecting sharp backing singers. They stand by the shelves and are small in the frame. Judge the clip by the lead singer.
Sunglasses or a mask. A hidden face comes out less like the person. Use a clear, front-facing photo.
A side-on or tilted photo. The model works best with a face that looks at the camera.
Thinking you need five photos. You do not. The lead photo alone gives you a clip with you at the microphone, and the other four performers stay as they are.
Expecting your own song. The clip comes with the template's music. Mouth movement follows the performance but will not match every word.

Posting It

The clip is vertical 9:16, so it fits TikTok, Reels and Shorts without cropping. It comes with the template's music, so you can post it as it is and add a caption about who plays what. If an app mutes the song, post the download without music and add the song from the app's own library. If you cast friends, tell them first and send them the file.

Timing helps. The trend took off as a college football joke posted right after a weekend of games, so the evening after your own game day, or your fantasy league's worst week, is a natural time to post one.

What You Get Back and How Long It Takes

The file is a 15-second MP4, 9:16, with an AI generated label, in two downloads: with the template's music and without it. It is usually ready about 5 minutes after payment, up to 40 when busy, and we email you when it is.

If a render fails on our side, a person checks it within 2 business days and we make it again or refund you in full. A finished video is kept for 30 days after delivery, so download it in that time. Prices: 480p $9.99, 720p $19.99, 1080p $39.99.

Is It a Filter?

No. A filter changes a video you filmed. We make a new 15-second clip from your photos with our video model: the people in your photos play their roles, and the roles you leave empty stay as they are.