How to Make a Gangnam Style AI Pet Video with Three Photos

Use one pet per image, with its face and characteristic features visible. For a dark-coated animal, choose lighting that separates the eyes and muzzle from the surrounding fur. For a pale animal, avoid a photo where bright light removes the detail around its nose, ears or coat.

Look for the details you use to recognize your own pet: an uneven facial patch, a white chest, ear shape or distinctive muzzle. Keep those features in the crop. A small animal photographed from across a room may be obvious to you from context, while leaving very little useful detail in the actual reference.

Choose a calm, clear moment rather than a motion-blurred action shot. You do not need to photograph your pet performing the movement; the template supplies it. Avoid placing a human hand across the face or including another animal that could confuse the reference.

Decide which pet gets each turn before uploading. If two animals look similar, use references with clearly visible distinguishing marks and double-check the thumbnails. The first file belongs in the center role, even though the first animal you see when reading a group picture from left to right is on the left.

Pet trio requires three photo slots. The one-selfie option is a separate mode designed for a human identity repeated three times; it is not the dedicated pet workflow.

1. Open the Gangnam Style studio and select Pet trio.
2. Add three JPEG, PNG or WebP photos, each no larger than 10 MB. Inspect every preview before continuing.
3. Choose 480p or 720p, check your current account or guest wallet balance, and confirm permission to use the photos. Follow any sign-in prompt shown.
4. Check the credit quote and submit once. Follow the task status until the prepared video is available.

A pet video costs 150 credits at 480p or 300 credits at 720p. Those are generation credits from the same Gangnam Style AI account used elsewhere on the site. Preparing a better photo before submission is more useful than treating repeated paid attempts as a substitute for a clear reference.

The output is a landscape 16:9 MP4 based on the 15-second reference performance; its finished duration can be slightly shorter. The reference soundtrack is added automatically. There is no separate audio upload or custom song selector in this workflow.

The intended pet treatment translates expressions and gestures into head and front-paw movement while retaining animal anatomy. That direction does not promise that every species, pose or transition will render equally well. Use the actual result to assess the movement rather than assuming the preview demonstrates an identical outcome for your pet.

Watch the opening group, each solo and the closing group. Check that the intended pet takes each turn, that prominent markings remain recognizable, and that paws, ears and faces move coherently. The quick camera changes can hide a mismatch when you only glance at the first frame.

If one pet is unclear, compare that result with its uploaded photo and change only the weak reference for a future attempt. For file or crop problems, use the photo checklist. If the task has failed or remains under review, follow the cost and fixes guide before creating another paid job.