FLUX
Generates high-resolution images with a rectified-flow transformer
WHAT IT IS
FLUX is a text-to-image model family Black Forest Labs released in August 2024. The team includes the original authors of Stable Diffusion, and FLUX.1 uses a rectified-flow transformer architecture with roughly 12 billion parameters in the main model. The family includes a closed pro version, an open-weight but non-commercial dev version, and a schnell version released under Apache 2.0. It supports text-to-image and image editing, and was widely adopted in the open image community after release.
Why it matters
Built by the original Stable Diffusion authors, it brought the rectified-flow transformer into open-weight image generation and became a major open-source text-to-image line in 2024.
Key specs
- Parameters
- About 12B in the main model
- Architecture
- Rectified-flow transformer
- Licence
- pro closed; dev open weights, non-commercial; schnell under Apache 2.0
- Released
- 2024-08
Capabilities
Related concepts
Diffusion Models
Learn a thousand tiny denoising steps, and you can build an image from pure noise
Latent Diffusion & Conditional Control
Run diffusion not over pixels, but inside a compressed semantic space
Multimodal Generation
One model that learns to speak, to draw, to move — even to model the 3D world
Comparable products
Stable Diffusion
2022Released text-to-image weights openly and small enough to run on consumer GPUs
Midjourney
2022A text-to-image service known for its aesthetic style
DALL·E 3
2023Rewrites a long prompt into a detailed description, then draws the image
Imagen
2022Generates photorealistic images with cascaded diffusion and a large text encoder
Seedream
2024A text-to-image model with native high resolution and strong text rendering