Skip to content

sd3.5

A lightweight inference model for image generation

Visit official site

What sd3.5 is

The model is described as a lightweight inference release with text encoders, a VAE decoder and core MM-DiT technology, made to help partner organisations implement SD3.5.

What you can do with it

  • Generate images from text prompts
  • Pick from several text encoders
  • Produce images at different sizes

Getting started

  • Download model files into the models folder
  • Set up a Python virtual environment
  • Run the inference script with a prompt

Pros & cons

✓ What we like

  • Inference is stated as efficient with low resource needs
  • Images of different sizes are supported

! What to watch out for

  • Usage follows the Stability AI Community License Agreement
  • Generation starts from text prompts and command-line settings

FAQ

What licence applies?

The material states the Stability AI Community License Agreement.

How is it run?

The steps describe running the inference script with a text prompt.

Last reviewed: 2026-11-15

More AI image generation tools

View all →

How we review