Ad
Skip to content
Subscribe Now
AI in practice
Copy the url to clipboard Share this article Go to comment section

Black Forest Labs expands FLUX.1 with four new AI tools for image editing

Matthias Bastian
Matthias Bastian View the LinkedIn Profile of Matthias Bastian
Nov 22, 2024
Image description
Black Forest Labs

Key Points

  • Black Forest Labs has released FLUX.1 Tools, a set of four new tools that provide more granular control and editing capabilities for images generated by the company's existing FLUX.1 image generation models.
  • The new tools include FLUX.1 Fill for inpainting and outpainting, FLUX.1 Depth and Canny for preserving image structure during complex transformations, and FLUX.1 Redux for creating image variations and restyling.
  • Black Forest Labs offers the tools commercially through its paid API and partners, as well as "Dev" versions with open-source code for developers and researchers.

Black Forest Labs has released FLUX.1 Tools, expanding its FLUX.1 text-to-image model with four new specialized AI tools.

The company says these tools provide more precise control over AI image creation and editing, and are available through both a paid API and open-source versions, depending on your usage scenario.

Four new AI tools for image processing

FLUX.1 Fill handles inpainting and outpainting tasks, letting users edit existing parts of images or extend them beyond their original borders. The company says the tool shows better performance than similar solutions in benchmarks.

Three-stage image transformation: business attire gradually becomes more casual, including neon sign change from
The series shows inpainting capabilities by transforming clothing styles from formal suits to casual denim jackets, with matching changes to the neon text. | Image: Black Forest Labs

With FLUX.1 Depth and FLUX.1 Canny, Black Forest Labs also introduced two structure-preserving tools. Depth uses depth information, while Canny relies on edge detection. These tools aim to maintain image structure during transformations, specifically for image re-texturing, with claimed better results than Midjourney's recent re-texturing feature.

Ad
DEC_D_Incontent-1

Two image sequences show the transformation process: the cave dweller and owl are transferred into different styles using depth maps/edge detection.
FLUX.1 Depth and Canny keep the original image structure intact during artistic changes. The depth maps and edge detection allow precise style variations while maintaining the same composition. | Image: Black Forest Labs

The fourth addition, FLUX.1 Redux, serves as an adapter for FLUX.1 base models, creating image variations through text prompts. It can create subtle changes to images or completely new styles while maintaining the core elements of an image.

Image sequence: minimalist still life with coffee cup and bread in different perspectives and lighting variations.
FLUX.1 Redux creates subtle variations of a coffee cup and pastry still life. The AI maintains the basic composition while adjusting perspective, lighting, and shadows based on text input. | Image: Black Forest Labs

Two versions for different needs

Black Forest Labs offers these tools through its paid API and as development versions with open-source inference code on Github and model weights on Hugging Face for developers and researchers. The tools will also be accessible through fal.ai, Replicate, Together.ai, Freepik, and krea.ai.

AI News Without the Hype – Curated by Humans

As a THE DECODER subscriber, you get ad-free reading, our weekly AI newsletter, the exclusive "AI Radar" Frontier Report 6× per year, access to comments, and our complete archive.

Source: Black Forest Labs