How to Make a Pet at the Window Video of Your Dog or Cat

Pet at the Window is the Window Peeper trend with your dog or cat in the lead. The camera drifts toward a lit window at night, your pet is stealing a bite of pizza, freezes when it sees the camera, tilts its head and pulls the curtain shut with a paw.

Your pet does not have to do anything. Upload one photo to the Pet at the Window generator and it renders a new 10-second clip.

Quick answer

Open Pet at the Window. The pet scene is already selected.
Upload one clear photo of one dog or cat, face and body visible.
Pick 720p or 480p and generate. It usually takes 2 to 6 minutes.
Download the 10-second, 9:16 video with sound and post it.

What does our test look like?

The cover of this guide is our own test: the AI-generated pet photo we uploaded on the left, a frame of the finished video on the right. This is the full result, made at Standard 480p:

Your pet will look different, and fur markings can shift a little. The example shows what the scene does, not an exact copy of your result.

What happens in the Pet at the Window video?

The scene runs a prepared prompt with fixed timing. The whole prompt is on the prompt page.

Time	What happens
0–2.5 s	The camera moves toward a lit window. Your pet is stealing a bite of a pizza slice on the table.
2.5–5 s	It bites the edge of the slice and chews, holding it steady with one paw.
5–7.5 s	It notices the camera and freezes, caught in the act.
7.5–8.5 s	It tilts its head toward the lens.
8.5–10 s	One paw pulls a heavy red curtain across the window. Only warm light shows behind it.

The sound is quiet night ambience, soft chewing, a paw rustle and the curtain swish, with no voices or music.

How do I make it step by step?

1. Choose a photo. One dog or cat, with its face and whole body in view. The photo is the only reference for its breed, fur and markings.

2. Pick a quality. High quality 720p uses a stronger model and gives a sharper picture, which helps with fur detail. Standard 480p is cheaper and fine for TikTok, Reels and Shorts.

3. Buy a pack if your balance is empty. Packs hold 1, 3 or 8 videos of one quality. Pay once, no subscription, and unused videos do not expire.

4. Generate. It usually takes 2 to 6 minutes. You can close the page; the video is saved in My Creations.

5. Check it and post. Watch it first, then download it and add your caption.

What pet photo works best?

One animal only. No second pet and no people in the photo.
Face and body visible, not only a close-up of the face.
Even light, so the fur colour and markings are clear.
No filter, text or sticker.
JPG, PNG or WebP, up to 10 MB.

Use a photo of your own pet, or one you have permission to use.

Does it work with any dog or cat?

The scene is made for one dog or one cat. It keeps the breed traits, fur colour, markings and body shape from your photo and does not turn the pet into a person or a cartoon. Unusual markings can come out simpler than in the photo; a sharper, evenly lit photo gives the model more to keep.

Why does my video look different from the example?

The markings are simpler. Spots and patches are the details AI video changes most. A clear photo and 720p help.
The head tilt is small. The prompt asks for a clear tilt, but the model can soften it.
The paw movement looks odd for a moment. Paws are hard for AI video. The prompt asks for natural paws, and most of the clip is your pet's face and body.

A failed generation returns its video to your balance automatically. A result you do not like is not a failure, so another try uses another video.

How much does it cost?

Quality	1 video	3 videos	8 videos
Standard 480p	$6.60	$16.60	$36.60
High quality 720p	$9.90	$24.90	$59.90

A video from a pack works with every scene of the same quality, so you can also make a Burger Thru Window video of yourself. Within 7 days of a purchase you can request a refund in Settings, minus the videos already made. Pricing has the details.