Gangnam Style AI is a browser-based AI video generator that maps 1, 2, or 3 uploaded portrait photos onto pre-choreographed Gangnam Style dance routines and exports a 15.0-second MP4 with the 132 BPM PSY audio track already merged into the file. The site is published by an independent editorial team (lead researcher Andy Swift) and is not affiliated with PSY or the original music video.
What is Gangnam Style AI?
Gangnam Style AI is a prompt-free web tool that converts static portraits into short dance videos. You upload JPG, PNG, or WebP images (up to 10 MB each) into labeled role slots, pick one of three fixed templates, choose 16:9 or 9:16 and 720p or 1080p, and receive a 24 fps MP4 with pre-synced 132 BPM audio. Every output is capped at exactly 15.0 seconds regardless of template.
What makes Gangnam Style AI stand out?
- Three fixed role-slot templates — Street Duo (2 photos, crossover position swap at 00:04), Terrace Trio (3 photos, opening wide shot plus Center, Left, and Right solo cuts), and Blue Tuxedo Solo (1 photo, which auto-adds a PSY-style blue tuxedo, bow tie, and sunglasses).
- Zero-prompt workflow — photographers drag photos into visual Role 1/2/3 zones instead of writing text prompts or shooting against a green screen.
- Face-lock claims — the vendor reports 98.4% facial identity retention across 450 frames, with decoupled cranial embeddings preventing face blending during position swaps.
- Pre-merged audio — the 132 BPM master stem ships inside the MP4, so footwork and lasso spins are timed to the beat without a separate editor.
- Smart 9:16 auto-reframe — a virtual camera pans to the active solo singer during the 04s–13s cuts and widens for the group finale.
- 3-second auto-refund — failed renders from server timeouts or pose-binding errors return all credits automatically.
- 24-hour privacy purge — uploads run in ephemeral sandboxes and are deleted within a day; the site states images are never used to train public models.
- Group photo splitter — one group shot can be detected and split into up to three separate role portraits on-device.
Who should use Gangnam Style AI?
- Friend groups and duos: combine three camera-roll portraits into the rooftop Terrace Trio routine for a birthday or group-chat joke.
- Office teams and pranksters: cast a deadpan boss in Role 1 and an energetic colleague in Role 2 for the Street Duo standoff.
- TikTok and Reels meme accounts: batch-render vertical 9:16 clips from supplied photos for short-form feeds.
- Pet or avatar owners: animate dogs, cats, toddlers, or stylized 2D anime characters through the bipedal skeletal mapper.
What can you do with Gangnam Style AI?
- Friend-trio tributes: upload three portraits, assign Center/Left/Right, and download a 15-second rooftop video with individual lip-sync close-ups.
- Solo novelty clips: drop one photo into Blue Tuxedo Solo mode and get a full-body horse-dance gallop in a PSY suit.
- Sports or team edits: the Soccer Trio example keeps each athlete's facial landmarks stable across quick camera changes.
How does Gangnam Style AI work?
- Select a template and assign portraits to role slots (about 15 seconds).
- An on-device Canvas pre-check verifies face angle and resolution in 0 ms, then you choose 720p (1 credit) or 1080p (2 credits) and toggle 9:16 auto-reframe.
- Cloud rendering runs 45–60 seconds and returns an unwatermarked MP4.
Pricing
Paid, one-time credit packs with no subscription: Starter at $9.90 for 2 credits, a 10-credit pack at $29.90, and a 24-credit pack at $59.90. One credit renders 720p, two render 1080p, and credits do not expire. Checkout runs through Stripe and supports Apple Pay and Google Pay.
Alternatives
Oakgen.ai (trio templates only, subscription-based credits) and APT Video AI (duo templates only, single-video purchase) are named on the site's comparison table.
FAQ
What photos work best?
Front-facing or 0–15° portraits cropped waist-up, at least 500×500 pixels per face, with even lighting. Side profiles beyond 45°, dark sunglasses, and hands over the mouth degrade landmark tracking and lip-sync.
How long are the generated videos?
Every template outputs exactly 15.0 seconds. Longer cuts and custom choreography are not offered.
Does the MP4 include the Gangnam Style music?
Yes. The 132 BPM PSY audio stem is pre-merged at 24 fps into every export, so the file is ready to post to TikTok, Reels, or WhatsApp without extra audio editing.
Is there a subscription?
No. Credits are bought one time and never expire; a failed render returns its credits within 3 seconds.
