The technology behind Flux builds on the advances of Stable Diffusion and takes image generation to a whole new level. Black Forest Labs, the company behind Flux, was founded by the creators of Stable Diffusion and has raised an impressive $31 million in a recently completed funding round. This development promises to fundamentally change the creative applications of AI and open up new possibilities for artists, designers and the creative industry as a whole.
Introduction of the FLUX.1 model family

FLUX.1 [pro]
The top of the FLUX.1 series, FLUX.1 [pro] delivers unrivalled performance in image generation. The model impresses with its outstanding prompt compliance, visual quality, image detail and output versatility. FLUX.1 [pro] is accessible for professional applications via the API. It is also available on Replicate and fal.ai.
FLUX.1 [dev]
FLUX.1 [dev] is aimed at non-commercial applications and is characterised by open weights and efficient, guidance-distilled performance. It is directly descended from FLUX.1 [pro] and offers similar quality and prompt following, but is more efficient than standard models of comparable size. The weights for FLUX.1 [dev] are available on Hugging Face and the model can be tested on Replicate or fal.ai.
FLUX.1 [fast]
FLUX.1 [fast] is designed for speed and local development and is ideal for personal use. It is freely available under the Apache 2.0 licence. Similar to FLUX.1 [dev], the weights are available on Hugging Face and the inference code can be found on GitHub and in the Hugging Face Diffusers. In addition, FLUX.1 [fast] offers a Day 1 integration with ComfyUI.
The vision behind Flux
For the founders of Black Forest Labs, Flux is more than just an image generation tool. Their vision is to revolutionise creative processes in various industries. Whether in art or design – Flux is intended to help creative people improve their work and realise their ideas faster and more efficiently. The focus is on user-friendliness, so that even users without in-depth technical knowledge can benefit from the advantages of AI technology.
FLUX.1 against Midjourney
Flux starts at the top with its three models and is therefore in direct competition with the large closed models such as Midjourney.
Is Flux a real contender against Midjourney in AI image generation?
— TechHalla (@techhalla) August 4, 2024
Let’s put them to the test with 7 prompts! 👇 pic.twitter.com/72fVBtSKG4
🤯 Is Midjourney in Trouble?
— Javi Lopez ⛩️ (@javilopen) August 3, 2024
Flux, created by the original team that launched Stable Diffusion and recently left Stability AI, is on par with Midjourney in terms of quality.
How is Midjourney going to react to this?
Here’s a link so you can try it yourself! pic.twitter.com/KzD2agpCLw
The discussions on X are in full swing
Will FLUX.1 become the new standard for text-to-image synthesis? Time will tell whether it can knock Midjourney off its throne, but one thing is certain: the future of AI-supported image generation is more exciting than ever.
Sources: