Turn your selfie into a festive Christmas portrait
Stage
Not given
Not given
Founded
Not given
Not given
Based in
US
GB
Pricing model
Not given
Paid
Pricing
Not given
One-time credit packs with no recurring subscription. A standard 1K generation uses one credit. See the pricing page for currently available credit pack details and costs.
Key features
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
Upload selfie in JPG, PNG or WEBP format up to 10 MB
No AI prompt writing required
Private uploads and creations
Preview finished image before downloading
One-time credit pack purchases with no subscription
Cost shown before generation begins
Output ratio options for portraits or social posts
What makes it different
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
No prompt needed - styles come prepared with a scene
Credits returned if generation fails due to confirmed technical failure
Linum
AIChristmas