FLUX

What is FLUX?

The FLUX repository from Black Forest Labs holds minimal Python inference code for its FLUX.1 open-weight image models. The lineup covers FLUX.1 schnell, dev and Krea dev for text-to-image, Fill for inpainting and outpainting, Canny and Depth models (including LoRA versions) for structural conditioning, Redux for image variation and Kontext dev for image editing. You clone the repo, install it in a Python 3.10 virtual environment, optionally add TensorRT support through an NVIDIA PyTorch image, and download weights from the linked Hugging Face pages. Licensing differs by model: schnell uses Apache 2.0, while the dev models use the FLUX.1-dev Non-Commercial License, so commercial use of those requires a paid license whose fee depends on monthly usage and is tracked through the BFL API. Black Forest Labs' hosted API, including non-open Pro models, is documented separately. The repository lists no prices.

Use cases

• Generate images from text locally
• Edit an image with Kontext
• Inpaint or outpaint part of a picture
• Guide generation with edges or depth
• Create variations of an existing image
• Track usage for commercial licensing
• Build image features on open weights

Who it's for

• ML engineers
• Python developers
• AI artists
• Researchers

Why use FLUX

• Runs the models on your own hardware, so prompts and images can stay local.
• Provides one permissively licensed model, schnell under Apache 2.0, alongside the dev models.
• Covers editing, variation and conditioning as well as plain text-to-image generation.
• Comes from the model's makers, which reduces guesswork around setup and usage tracking.
• Points to a hosted API for people who prefer not to run models themselves.

FLUX features

• FLUX.1 schnell, dev and Krea dev text-to-image
• FLUX.1 Fill for in-painting and out-painting
• FLUX.1 Canny and Depth structural conditioning
• Canny and Depth LoRA variants
• FLUX.1 Redux for image variation
• FLUX.1 Kontext dev for image editing
• Optional TensorRT-accelerated installation
• Built-in usage tracking through the BFL API
• Weights hosted on Hugging Face

FLUX integrations

• Hugging Face
• NVIDIA TensorRT
• PyTorch

How to get started with FLUX

1. Clone the repository and create a Python 3.10 virtual environment.
2. Install the package with pip, or follow the TensorRT instructions using the NVIDIA PyTorch image.
3. Download the model weights you want from the linked Hugging Face pages.
4. Run the CLI or demos with a prompt or an input image, adding track_usage for commercial use of open weights.
5. Check each model's license before using the outputs commercially.

FLUX limitations

• Needs a capable local GPU and your own Python environment
• FLUX.1 dev models use a non-commercial license
• Commercial use of open weights other than schnell needs a paid license billed on monthly usage
• The repository is minimal reference code rather than a finished app

FLUX pricing

FLUX is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

FLUX alternatives

The main FLUX alternatives listed on Debuly are Midjourney, Ideogram, Adobe Firefly, Recraft, HiDream.ai and Black Forest Labs FLUX. The table compares pricing and platforms; order is A–Z.

1. 

   Adobe Firefly

   Adobe's generative AI app for creating and editing images, video, audio and vectors, with Adobe and partner models.

   Image generationVideoDesignAudio and voice
   FreemiumVisit Adobe Firefly (official site)
2. 

   Black Forest Labs FLUX

   FLUX image and video models from Black Forest Labs, sold by API credits or as licensed open weights you can fine-tune and self-host.

   Image generationVideoDeveloper platforms
   PaidVisit Black Forest Labs FLUX (official site)
3. 

   HiDream.ai

   AI image and video generation platform with open-weight image models, creator tools and an enterprise API service.

   Image generationVideoDeveloper platforms
   Visit HiDream.ai (official site)
4. 

   Ideogram

   AI image and video generator known for readable text in images, with editing tools, batch generation, an API and credit-based plans.

   Image generationDesignVideo
   FreemiumVisit Ideogram (official site)
5. 

   Midjourney

   AI image and video generator that turns text prompts into artwork, sold as monthly or annual subscriptions with no free plan.

   Image generationVideoDesign
   PaidVisit Midjourney (official site)
6. 

   Recraft

   AI image and design platform that generates raster and vector images, mockups and brand styles, with editing tools and an API.

   Image generationDesignMarketing and SEO
   FreemiumVisit Recraft (official site)

FLUX FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from avatars.githubusercontent.com · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.