Free: one 1K watermarked image per day. Credit packs: Starter $4.99 for 50 credits, Basic $8.99 for 120 credits, Standard $17.99 for 260 credits. One watermark-free image costs 1 credit with default settings.
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
Combine two or more reference images
Multiple AI models available (FLUX, Nano Banana, Seedream, Qwen, GPT Image, Grok, Wan)
Text description guidance for composition
Up to 4K output resolution on supported models
Free daily generation without signup
Watermark-free generation with credits
No editing software required, browser-based
Support for JPEG, PNG and WebP upload formats
Task history and image preview for 7 days
Adjustable output aspect ratio and format
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
Free daily generation with no signup required
Multiple advanced AI models in one interface
One-time credit purchases with no expiry
Browser-based with no software installation needed
Linum
AI Image Combiner