Tuesday, September 27

Local source: you can now run Stable Diffusion AI art generation on your Mac M1

It has a one-click installer, runs locally, requires “no dependencies or technical knowledge” and, most importantly, does not communicate with the cloud. Diffusion Bee is a simple way to run all that AI art stuff on your Mac M1 (or M2).

There really is a lot more to be said about what stable streaming and similar generative machine learning art models are and what they could mean. But I think it’s worth noting that now you can get this to work locally. This means that your work is not subject to any data collection – not to mention that it’s quite fun to do it locally.

In fact, whether this machine learning app is your cup of tea or not, it’s exactly the kind of potential Apple Silicon promises. While you can do a lot of work with PCs and NVIDIA GPUs, what the Mac has to offer is simplicity and a single vendor. And, to put it another way, while there are some great PCs out there, the laptop landscape is more of a mixed bag. The Mac has the ability to do a lot while remaining cool and quiet.

And this “one-click” “dependency-free” thing is clearly appealing.

Features, from the dev:

Complete data privacy – nothing is sent to the cloud

Clean and easy to use user interface

One-click installer

No dependencies needed

Multiple image sizes

Optimized for M1/M2 chips

Runs locally on your computer

Check out the GitHub:


Before there’s a misunderstanding, “local” always implies Stable Diffusion checkpoints and datasources. You are not using your own data as a training set. This is the main caveat of this whole supposed AI art phenomenon. The AI ​​doesn’t “do” the art here; it relies on very large and very particular data sets, which also determine the outputs you get.

I’m personally interested in their other tool, Liner.ai, which works with your own training data – and is also easy to set up on your local machine. This also works on Windows and Linux:


It’s not just that controlling the dataset is more ethical; this is perhaps the most interesting long-term application of machine learning. This can mean material that is both more personal and varied.

In the short term, however, Stable Diffusion and its ilk are both fascinating and worthy of critical analysis.