Turn three photos into a 19-second Gangnam Style AI video of you and two friends
Generative video models for animation
Stage
MVP
Not given
Founded
2026
Not given
Based in
HK
US
Pricing model
Paid
Not given
Pricing
Standard 480p: 100 credits ($9.99 for single pack, $19.98 for 210 credits, $39.96 for 440 credits); HD 720p: 220 credits. No subscription, no free tier.
Not given
Key features
Generates 19-second landscape video (16:9 format)
Accepts three photos (separate or one group photo)
City street setting with night lighting
Individual solo close-ups for each person
Timed to song's 132 BPM beat
No prompt required
Browser-based, no app installation
Video comes without audio
English and Spanish language support
Pyramid-JiT (P-JiT) architecture with 32×32 token reduction
4× more aggressive token reduction than Linum v2
Converges in 11× fewer samples than Linum v2
2B-parameter model with frozen Qwen text encoder
Pixel-space diffusion with multi-resolution readouts
PixelREPA alignment with DINOv3-L
Curriculum learning over noise schedule
What makes it different
Built specifically for one trend with no customization needed
One-time credit purchase with no subscription
Automatic refund if video render fails
Photos deleted after 7 days, never used for training
Low-cost training: full 2B model trained for just $21K in compute
Controls baked into model training rather than post-hoc fine-tuning
Pixel-space compression without separate VAE bottleneck
Apache 2.0 open release with full model weights and code
Gangnam Style AI Video Generator
Linum