How to Make the First Time in a Fancy Cafe AI Video from a Photo

A bare concrete cafe. Someone balanced on a tiny metal disc fixed to the wall, coffee in hand, looking like a regular. Then a stranger walks in, sits on the bench and puts their cup on the identical disc. That's the "POV: your first time in a fancy cafe" trend, and here's how to make it with your own face, from one or two photos.

What is the First Time in a Fancy Cafe trend?

It's a POV joke that took off on Instagram and TikTok around 31 October 2026. Someone in a minimalist cafe sits on a little round wall-mounted table, thinking it's a stool. When another customer uses the same thing as a table, they quietly climb down and act like nothing happened. Creators rushed to remake it, and many used AI video tools to put themselves in the scene.

How to make it

1. Open the create form.
2. Add one or two photos. One person per photo, face sharp and forward, in good light. Photo 1 sits on the table and photo 2 walks in; with one photo, a made-up stranger walks in.
3. Pick the model, length and quality. Budget and Balanced render 15 seconds with the whole gag; Premium also renders the whole 21 seconds, up to 1080p.
4. Generate. Your photo is reviewed first, then the video renders in a few minutes. You can leave the page: the link is emailed when it's ready.
5. Download the MP4. It comes with the trend's sound. To pick the sound yourself when you post, turn the song off or download without music.
6. Post it to TikTok, Reels or Shorts with the caption "POV: your first time in a fancy cafe". The video itself has no text on it.

What happens in the video

• 0–5.75 s: the perch. You sit on the little wall table, knees up, take a long sip and look around like a regular.
• 5.75–9 s: the entrance. A stranger strolls in with a coffee, passes in front of you and sits down on the bench.
• 9–14.75 s: the side-eye. Their cup goes on the identical table. You glance over, glance down at what you're sitting on, and slowly get it.
• 14.75–17.5 s: the climb down. An awkward scramble onto the bench, then your cup goes on the table as if that was always the plan.
• 17.5–21 s: nothing happened. Both seated, both cups on their tables, you staring straight ahead.

The 15-second cut runs from 3.5 to 18.5 s, from just before the entrance to the moment you sit down, so the joke stays whole. The camera never moves, and the video is vertical (9:16).

Photos that work best

• Face clearly visible. The side-eye is the whole joke, so the face has to read. No sunglasses, masks or heavy filters.
• Good light. Daylight or a bright room.
• Waist up or more. You're on screen head to toe, so a fuller photo gets your build and clothes right.
• One person per photo. Crop group shots to the person you mean.

Your clothes come from your photo. We don't dress you in the clip's outfits, and the cafe, the tables and the coffee are always generated, so your background doesn't matter.

Who walks in?

With one photo, a well-dressed made-up stranger does. Add a second photo to cast a friend, partner or sibling who agreed to it. Adults only.

What does it cost?

One video starts at $6.99 at checkout with no account, and the whole 21 seconds on Premium at 480p is $9.99. With credits, the whole take is 25 credits at 480p; credit packs start at $9.90 for 20 credits, and credits never expire. No watermark. If a render fails or a photo is rejected, credits come back automatically, or a single video can be made again for free.

FAQ

Do I need to film anything? No. Any clear photo works. The cafe, the table and the coffee are generated.

Can I use a photo of my dog or cat? Yes, with the dog version or the cat version.

What's the song? "Oh Yeah?" by Steve Lacy, the title track of his 2026 album.

Can I use a celebrity's photo? No. Celebrity photos and film stills don't pass review. Use photos of yourself or people who agreed to appear.

Make your Fancy Cafe video →