← Back to glossary

Stable Diffusion

An open-source neural model that generates images from a text prompt. Morf Vision uses it to render interiors.

In depth

Stable Diffusion is a latent diffusion model released by Stability AI in August 2022. It produces images from a text prompt (text-to-image) or transforms an existing image (image-to-image). Unlike DALL·E or Midjourney it is open-source and can be fine-tuned for narrow domains. Morf Vision runs Stable Diffusion XL with an in-house LoRA fine-tune trained on roughly two million interiors, which is what lets it keep furniture proportions and room geometry believable instead of hallucinating impossible spaces.

Examples

  • 1.A photoreal interior render produced in 30–60 seconds
  • 2.Recoloring the walls in a finished render without redrawing the scene
  • 3.Generating reference imagery from a written brief

Morf Vision is a platform that automates interior design using the technologies described above.

Try it free