If you want to try out Stable diffusion, which we would like to introduce in our own article, you don’t have to train a neuronal network right away. Apple represents a fully converted Model in Core ML format Ready that can be used directly – including support for CPU, GPU and Neural Engine.
Our practical example developed here uses prepared models and thus already creates the first pictures by command line. It continues to compile an app for iOS and MacOS from Apple’s sample projects for Swift and Xcode. This creates a colorful result from the gray theory of the previous page in practice, which is literally impressive.
Apple’s frameworks for training and executing models are called Create ML and Core Ml. Core ML is enough for the example, as it is already trained Stable diffusion model from Hugging Face Used and Create ML therefore does not require. The portal is a collection point for models, libraries and tools in the context of artificial intelligence.
That was the reading sample of our Heise Plus article “MacOS: Stable Diffusion integrating into your own app”. With a Heise Plus subscription you can read the whole article.
Discover more from Apple News
Subscribe to get the latest posts sent to your email.