One photo is all it takes
Upload a single JPG, PNG or WebP up to 10 MB. You, a friend who agreed, or your pet becomes the one caught eating in the window. One face or one pet per photo works best.
Upload one photo of yourself, a friend who said yes, or your pet. You get a vertical 720p MP4 of 6, 10 or 15 seconds: at night the camera creeps toward the one lit window, you’re caught mid-burger, you stare into the lens, then you pull the curtain shut. No watermark, ready to post.
1Your photo
Upload one photoJPG, PNG or WebP · up to 10 MBOne face or one pet, facing the camera in good light; head-and-shoulders or half-body is perfect. No group shots, celebrities or screenshots with another app’s watermark.
2Video length
Continue with GoogleTrending on TikTok & CapCut
Clips are creators’ TikTok posts, credited to them. None of them was made on this site.
How the scene plays
The camera starts across the street from a dark apartment block and slowly pushes in on the only lit window. Inside, the main character sits alone at a small table with a big burger. Mid-chew they notice they’re being filmed, freeze, and stare straight into the lens; the stare is the longest beat. Then the burger goes down, they stand up and pull the curtain shut. No words, no on-screen text: the curtain is the punchline. It’s AI, so no two renders are identical: the likeness, hands or paws, the timing of each beat, the room details and the sound all vary.
Make My Window Peeper VideoNo template hunting, no editing app, no timeline. One photo goes in and the whole window scene comes out, ready for your feed.
Upload a single JPG, PNG or WebP up to 10 MB. You, a friend who agreed, or your pet becomes the one caught eating in the window. One face or one pet per photo works best.
Pets get the full treatment: your dog or cat sits upright at the table, holds the burger in its paws, freezes when it spots the camera and stares you down before the curtain closes.
Every length has the whole sequence: the zoom, the bite, the stare, standing up and the curtain. 6 seconds costs 6 credits, 10 seconds costs 13 and 15 seconds costs 20. The 15-second version is the longest, close to the original template, and gives each beat the most time.
You get a clean 720p MP4 in 9:16, the shape TikTok, Reels and Shorts expect. Download it, or tap “Share to TikTok”: the caption is copied and your device’s share menu opens with the video, so you pick TikTok and check the post there. Every video stays saved in your account.
The AI adds quiet city ambience, chewing and a paper-bag rustle, a low music sting when you’re spotted and the swish of the curtain. There’s no speech or lyrics, so you can add the trend sound on top when you post.
If a render fails we retry it once for free. If it fails again, your credits come back automatically. Not happy with a finished video? Email support and we’ll refund your payment in money, not credits.
The classic version: you, alone at a little table with a big burger, and then you realize someone across the street is filming. Pick a front-facing photo where your face is clear and well lit; head-and-shoulders or half-body both work. The stare is the longest beat of the clip, so your face is the joke. Want to star a friend instead? Use their photo only if they said yes.
Pets in the window are one of the big variations of this trend. Upload a clear photo of your dog or cat, one pet per photo, ideally looking toward the camera. The AI sits them upright at the table with the burger in their paws, then gives them the same slow zoom, the frozen mid-bite stare and the curtain at the end.
Creators online have remixed the window with a kitchen at night and other snacks. This generator always makes the same night burger-window scene: you can’t choose the room, the food or the actions. The room is generated fresh each time from a written description of the scene rather than copied from the original clip, and the model sometimes keeps parts of your photo’s own look, so its details change from video to video. What stays is the order: the zoom, the bite, the stare, the curtain.
A 6-second video costs 6 credits, 10 seconds costs 13 credits and 15 seconds costs 20 credits. Every length has the whole scene; the 15-second one gives each beat the most time. Packs are one-time payments in USD and the credits never expire. There are no free videos; signing up with Google is free.
Single
One 15-second video to try the trend
Trio
You, your best friend and your dog
Crew
Five 15-second videos for the group chat
FAQ
Can’t find your answer here? Email [email protected] and we’ll reply by email.
It’s an AI video trend: at night a camera across the street slowly zooms in on one lit apartment window, where someone is secretly eating a burger alone. They notice they’re being filmed, stop chewing, stare straight into the lens, then get up and pull the curtain. It began as an AI-made concept ad posted on Instagram by @karina.aicontent on 11 October 2026 and blew up as a CapCut template on TikTok from 20 October. Window Peeper AI makes the same kind of scene with your own photo in the window.
No. Signing up with Google is free, but every video uses credits and there are no free videos. The smallest pack, Single, is $9.90 for 20 credits: one 15-second video or three 6-second ones. Credits are a one-time purchase and never expire.
You don’t lose your credits. A failed render is retried once automatically for free, and if it fails again the credits go straight back to your balance. Photos refused by the content check aren’t retried, but those credits come back too. If a finished video isn’t what you hoped for, email [email protected] and we’ll refund your payment in money, not credits.
Your photo is used only to make your video. It goes to our AI video provider for that render, and the provider doesn’t keep it; the finished video is saved to your account, not posted anywhere. Email [email protected] at any time to have your photos and videos deleted.
Yes. Upload a clear photo of one pet and it becomes the one in the window: it sits upright at the table holding the burger in its paws, freezes when it notices the camera, stares into the lens and then the curtain shuts. One pet per photo, facing the camera and in good light, gives the best result.
No, the trend’s TikTok sound isn’t included. Your video has AI-made sound instead: quiet night city ambience, chewing and a paper-bag rustle, a low music sting when you’re spotted and the curtain swish, with no speech or lyrics. The trend sound is labelled “original sound – rodentspawn”; add it from TikTok’s sound library when you post, and turn your video’s own sound down in TikTok’s volume settings if you only want the trend sound.
Yes, CapCut has a template called “Window Peeper”; its page showed 60.69K uses on 23 October 2026, a number that keeps changing. You don’t need CapCut or any app here. Window Peeper AI works in your browser: you upload one photo and get the finished MP4 back.
You choose 6, 10 or 15 seconds. Every length has the whole scene, from the zoom to the curtain; 15 seconds is the longest, close to the original template’s 14 seconds, and gives each beat the most time. A render takes about 2 minutes for 6 seconds, about 3 minutes for 10 and about 4 minutes for 15; these are estimates, and it can take longer when the queue is busy. You can leave the page while it works, because the video is saved to your account.
One photo of you, a friend who’s in on it, or your pet. Pick a length, and a few minutes later you’ll have the zoom, the stare and the curtain, ready to post.