Skip to content

FLUX.1-schnell

A 12-billion-parameter rectified flow transformer for text-to-image

Visit official site

What FLUX.1-schnell is

The model is described as a 12-billion-parameter rectified flow transformer that generates images from text descriptions, trained with latent adversarial diffusion distillation and able to generate images in 1 to 4 steps.

What you can do with it

  • Generate images from text descriptions
  • Use latent adversarial diffusion distillation
  • Produce images in 1 to 4 steps
  • Reach the model through an API and Comfy UI

Application scenarios

  • Create digital artwork
  • Support environment and character design in game projects
  • Produce visual content quickly for advertising and marketing

Getting started

  • Visit the GitHub repository for the reference implementation and sampling code
  • Set up a development environment from the provided code and docs
  • Run local inference through an API endpoint or Comfy UI
  • Enter a text description to generate an image

Pros & cons

✓ What we like

  • Images can be generated in 1 to 4 steps
  • A reference implementation, sampling code, an API and Comfy UI access are listed

! What to watch out for

  • Local inference needs a development environment set up from the code and docs
  • Access for local inference goes through an API endpoint or Comfy UI

FAQ

How large is the model?

The material states 12 billion parameters in a rectified flow transformer.

What license applies?

Apache-2.0 is named, for personal, scientific and commercial purposes.

Last reviewed: 2026-11-15

More AI image generation tools

View all →

How we review