Skip to main content

BlackForestLabs Introduces FLUX.1 Tools

BlackForestLabs Introduces FLUX.1 Tools

BlackForestLabs (BFL) has announced the launch of FLUX.1 Tools, an advanced suite of models designed to enhance control and steerability in text-to-image generation. Building on the capabilities of the FLUX.1 base model, these tools enable users to modify and re-create both real and generated images with remarkable precision.

The newly released tools are part of the FLUX.1 [dev] open-access model series and the BFL API, which includes professional-grade variants. This comprehensive suite features four key components:

  • FLUX.1 Fill: A state-of-the-art inpainting and outpainting model for seamless image editing.
  • FLUX.1 Depth: A model for structural guidance using depth maps extracted from input images.
  • FLUX.1 Canny: A model leveraging canny edge detection for structural preservation in image transformations.
  • FLUX.1 Redux: An adapter for generating image variations and restyling.

This launch reaffirms BFL’s dual commitment to delivering cutting-edge, open-weight models for researchers and offering premium tools for developers and businesses. FLUX.1 Tools are also being integrated into platforms such as fal.ai, Replicate, Together.ai, Freepik, and krea.ai.


Inpainting and Outpainting with FLUX.1 Fill

The FLUX.1 Fill model brings advanced inpainting capabilities that surpass existing tools, including Ideogram 2.0 and open-source solutions such as AlimamaCreative’s FLUX-Controlnet-Inpainting. It also offers outpainting features, enabling users to extend images beyond their original borders.

Benchmark results indicate that FLUX.1 Fill [pro] outperforms all competing models, establishing itself as the industry leader. Even the open-access FLUX.1 Fill [dev] variant offers superior efficiency and performance compared to proprietary alternatives.


Structural Conditioning with FLUX.1 Canny / Depth

The FLUX.1 Depth and FLUX.1 Canny models leverage structural conditioning to ensure precise control during image editing. By utilizing depth maps or canny edge detection, users can preserve the original structure of images while applying text-guided edits.

According to benchmarks, FLUX.1 Depth [pro] offers unmatched diversity and consistency in depth-aware tasks, while FLUX.1 Canny [pro] leads in edge-based editing. Open-access variants are also available under the FLUX Dev License.


Image Variation and Restyling with FLUX.1 Redux

FLUX.1 Redux is designed for image variation and restyling, providing users with the ability to refine or create slight variations of existing images. When combined with the latest FLUX1.1 [pro] Ultra model, Redux supports high-quality 4-megapixel outputs and flexible aspect ratios.

Benchmarks show that Redux delivers state-of-the-art performance, positioning it as a leading tool for image variation workflows.


Partnerships and Ecosystem Expansion

With the introduction of FLUX.1 Tools, BlackForestLabs aims to expand the vibrant FLUX ecosystem. The tools are expected to significantly enhance the capabilities of researchers and developers, further solidifying FLUX.1 as a leading platform for text-to-image generation and editing.

For more information and to access the API, visit api.bfl.ml.

Share

AD

You may also like

0
    0
    Your Cart
    Your cart is emptyReturn to Courses