Black Forest Labs models and APIs

About Black Forest Labs

Black Forest Labs is represented here by FLUX.1 [dev], its 12-billion-parameter open-weights image model, offered as a LoRA family: you bring the weights, the platform runs them.

FLUX.1 Dev LoRA publishes text-to-image. Each request loads up to three LoRA files trained for FLUX.1 [dev], each with its own strength from 0 to 4, and returns one image at about one megapixel in any of nine aspect ratios, with a guidance control from 0 to 20. More than a thousand community LoRAs on Hugging Face are built for this base, from film realism and editorial finishes to cartoon styles and characters, and every image costs the same whatever the ratio.

Come here when the look you want already exists as a FLUX LoRA: paste the file link, set the scale and put its trigger phrase first. This family makes new images only; to change a picture you already have, use an image family that publishes editing. The platform adds no content filter, though in our tests FLUX held part of the scene back at the strongest content level.