PixArt-Sigma XL 2 1024MS

Conditional

PixArt / Huawei Noah's Ark Lab · PixArt · Released 2024-03

0.6BImage

A remarkably small diffusion transformer that reaches 4K-capable text-to-image quality at a training cost far below SDXL-class models.

Strengths

  • +Very small parameter count for the quality achieved
  • +Low training and fine-tuning cost
  • +Runs easily on modest consumer GPUs

Limitations

  • -Smaller community ecosystem than Stable Diffusion or FLUX
  • -OpenRAIL++ license carries use-based restrictions

License

CreativeML Open RAIL++-M

free with acceptable-use or scale conditions - read it

Hardware

runs on a good consumer GPU (or Apple Silicon) with quantization

Links

Stats

PixArt-alpha/PixArt-Sigma-XL-2-1024-MS— downloads·— likes

via Hugging Face

Related models