StableML Unveils StableGUI: A Generative Model for Pixel-Perfect UI Mockups

AI · 5 min read

StableML released StableGUI, a model designed to generate pixel-perfect UI mockups by conditioning on component libraries, color palettes, and textual briefs. Unlike general image models, StableGUI enforces component boundaries and alignment heuristics to produce usable interface images.

The model supports constrained generation, where designers can pin elements or provide masks to restrict where new elements are placed. StableML also included a refinement pipeline that converts outputs into layered PSD or Figma-compatible files with editable components.

StableML emphasized transparency: StableGUI publishes its training dataset composition and includes a human-in-the-loop mode for iterative approval—helpful for teams that need to maintain brand fidelity while leveraging generative exploration.