Skip to main content

Improve Lustify SDXL training set

Feature RequestsImage

Generated images are very biased to m/f. Widen training for solo, f/f subjects. Fluids also currently very limited. Anatomy is usually inaccurate currently.

This might be achieved by adding endgame checkpoint or later.

Status: Rejected1 comment

Log in to comment and vote

Comments1

  • Gray Bauble

    •

    Aug 6, 2025

    Venice does not train the models ourselves, we rely on open source providers of these models.