[13:55 Fri,22.November 2024 by Rudi Schmidts] |
The German AI company Black Forest Labs has released new Flux Tools, designed to make their text-to-image model FLUX.1 more accessible to everyone. The release of the FLUX.1 Tools consists of four distinct features, available as freely accessible models within the FLUX.1[dev] series and in the BFL API as an extension to FLUX.1[pro]:
![]() FLUX.1 Depth: Models trained to preserve structure based on a depth mask extracted from an input image and a text prompt. FLUX.1 Canny: Models trained to preserve structure based on extracted edges derived from an input image and a text prompt. Both variants retain the original image structure through edge or depth masks, enabling users to make text-guided changes while keeping the core composition intact. This is particularly effective for retexturing images. ![]() FLUX.1 Redux: A model that allows mixing and recreating input images and text prompts. For a given input image, FLUX.1 Redux can reproduce the image with slight variations, refining it as needed. ![]() Each tool is available in the BFL API as a FLUX.1[pro] variant, while freely accessible FLUX.1[dev] versions with inference code and weights are also offered. Additionally, the models can be accessed via web services such as fal.ai, Replicate, Together.ai, Freepik, and krea.ai. ![]() deutsche Version dieser Seite: Black Forest Labs Flux Tools machen generative KI einfach(er) nutzbar |
![]() |