stability-ai / sdxl

A text-to-image generative AI model that creates beautiful images

  • Public
  • 70M runs
  • GitHub
  • Weights
  • Paper
  • License
  1. Author
    @daanelson

    7762fd07

    Latest
  2. Author
    @daanelson

    This version has been automatically upgraded to a newer version.

  3. This version has been automatically upgraded to a newer version.

  4. This version has been automatically upgraded to a newer version.

    Release notes

    • fine-tuned weight loading improvements
  5. Author
    @daanelson

    This version has been automatically upgraded to a newer version.

    Release notes

    • Image decoding bugfix
  6. Author
    @daanelson

    This version has been automatically upgraded to a newer version.

    Release notes

    • Even faster cold boots
  7. Author
    @daanelson

    This version has been automatically upgraded to a newer version.

    Release notes

    • Enables providing your own captions for training (as opposed to generating them)
  8. Author
    @anotherjesse

    This version has been automatically upgraded to a newer version.

    Release notes

  9. Author
    @anotherjesse
    Version
    22.04

    This version has been automatically upgraded to a newer version.

    Release notes

    • Inpainting now respects width/height.
    • Fix ROI problem when training.
    • BLIP caption_prefix now does not interfere with BLIP captioner when training.
  10. Author
    @cloneofsimo
    Version
    22.04

    This version has been automatically upgraded to a newer version.

  11. Author
    @cloneofsimo
    Version
    22.04

    This version has been automatically upgraded to a newer version.

  12. Version
    22.04

    This version has been automatically upgraded to a newer version.

  13. Author
    @daanelson
    Version
    22.04

    This version has been automatically upgraded to a newer version.

  14. Author
    @daanelson
    Version
    22.04

    This version has been automatically upgraded to a newer version.