Skip to content

stable-diffusion-3.5-large-turbo

Fast text-to-image generation with fewer steps

Visit official site

What Stable Diffusion 3.5 Large Turbo is

The model is described as a multimodal diffusion transformer that generates images from text with adversarial diffusion distillation. It targets better quality and fewer inference steps.

What you can do with it

  • Generate images from text prompts
  • Reduce the number of inference steps
  • Use a quantised model on low VRAM GPUs

Model notes

  • QK normalisation stabilises training
  • Terms must be accepted on Hugging Face
  • A Stability Community License applies

Pros & cons

✓ What we like

  • Inference step count can be customised
  • Conditional generation follows text prompt styles

! What to watch out for

  • Use is governed by the Stability Community License
  • Free use is limited to research, non-commercial work and entities under 1M USD annual revenue

FAQ

What technique speeds it up?

Adversarial diffusion distillation, or ADD.

Where is it published?

On Hugging Face under the Stability Community License.

Last reviewed: 2026-11-15

More AI image generation tools

View all →

How we review