New AI products and services can appear without warning, like a bolt of lightning, or they can appear after months of continuous promotion. Flux by startup Black Forest Labs is the former.

This AI image generation model has been called a legitimate successor to Stable Diffusion, and after its release it was compared directly to market leader Midjourney and quickly spread.

The difference between Flux and Midjourney is that Flux is open source and runs on laptops with modest performance. In other words, like Stable Diffusion, it is or will be available on many of the multi-model platforms like Poe, Nightcafe, and FreePik.

I have been using it and my initial impression is that it is better than Midjourney in some areas, especially for rendering people, but its skin textures are not as good as Midjourney v6.1.

Flux was born out of the AI startup Black Forest Labs. This new company was founded by the people responsible for most modern AI image generation technology.

The Germany-based company is led by Robin Lombach, a former engineer at Stability AI, Andreas Bratman, Dominik Lorenz, and other key figures in diffusion-based AI model development. This is also the technology that powers many AI video tools.

Currently, there are three versions of Flux.01, all of which are text-to-image models: the first is a commercially licensed Pro version, used primarily by companies like FreePik to provide access to generative AI image technology to their subscribers; the second is a Pro version, which is used by companies like Pik to provide their subscribers with access to generative AI image technology.

The next two are Dev and Schnell. These are mid-weight, fast models that outperformed Midjourney, DALL-E, and Ideogram in prompt compliance, image quality, and text rendering on images in my tests run on a laptop with RTX 4090.

The company is also working on a text-to-video model that promises to deliver high-quality output and be available as open source. Branding "cutting edge text-to-video for all."

If you have a well-equipped laptop, you can download Flux.01 and run it locally. There are simple ways to do this, such as using the Pinokio launcher, which allows you to install and run the AI model with a few clicks, and it is free. However, the file size is large.

But if your machine is not up to the task, there are already several websites that offer access to Flux.01, which may contain the largest commercial Pro models.

One of my favorite AI imaging platforms, NightCafe, already has access to this model and can immediately compare it to images from other tools such as Ideogram and Stable Diffusion 3.

Poe, an AI model platform, has access to Flux.01 and can generate images in a chatbot-style format similar to the way images are created using tools like ChatGPT or DALL-E.

It can also be accessed through more popular developer platforms such as Based Labs, Hugging Face, and Fal.ai. FreePik, one of the largest AI image platforms on the market, has stated that it is also working on bringing Flux to its own site.