FLUX is a family of image generation models from Black Forest Labs, offered both through a paid API and as open-weight models you can download.
In plain words
FLUX.1 [dev], one of the open versions, is a 12-billion-parameter rectified flow transformer: it learns a direct path from random noise to an image that matches the text prompt. It was trained with guidance distillation to make generation more efficient. Its weights are open for research and art, while commercial use of the model itself needs a licence. FLUX models can also edit images, and FLUX 3 covers video and audio too.
A quick example
Imagine a designer who wants a poster mock-up. They could run FLUX.1 [dev] locally in ComfyUI, type a prompt such as "retro travel poster of a mountain lake", and get an image back.
Where you'll meet it
- Image generation and editing through an API or locally.
- Workflows in ComfyUI and Diffusers.
Related
Sources
- DocsFLUX.1 [dev], Black Forest Labs (Hugging Face) · read 29 Sept 2026
- Docsflux, Black Forest Labs (GitHub) · read 29 Sept 2026
- DocsBlack Forest Labs, Black Forest Labs · read 1 Oct 2026
Checked by WhatIsAI against the sources above · 1 Oct 2026. Spotted an error? Report it.