Running Stable Diffusion on Android

Stable Diffusion is the latest open-source text-to-image neural network. It builds on well-established neural architectures and improves them by introducing image embeddings, allowing the networks to be trained more efficiently, and allowing for better outputs whilst all being computationally cheaper.

Stable Diffusion has been ground-breaking for many reasons, but perhaps the most important is the fact that the model is open source. This means that anyone can download it, modify it, and run it on their own machines. The efficiency improvements that Stable Diffusion has achieved not only mean that the neural network is easier to train, but also less demanding to run. has brought Stable Diffusion to Android for free:

The knock-on of StabilityAI’s efficiency improvements is that consumer GPUs can now run Stable Diffusion! This is an amazing leap forward for the AI industry.

HOWEVER, people who do not own powerful modern machines still will not be able to run Stable Diffusion, which still means that using Stable Diffusion has an entry barrier.

To help solve this, has created a free android app which allows users to run Stable Diffusion on servers hosted by Microsoft. This means that now anybody with an Android phone can access Stable Diffusion for free, whenever they want! I am thinking about creating this for IOS too. Please leave a comment if you are interested!

The app can be found on the app store under the name “Make AI Art”.

The app is still new and might contain bugs. Please don’t be too harsh on it! It costs a decent amount of money to host!

If you do enjoy the app, please also consider leaving it a nice Google Play rating! Most people only leave reviews when something goes wrong, so positive feedback is always welcome!