Apple Improves Mac, iPhone, iPad Software To Accommodate AI Art-Making
By Mikelle Leow, 02 Dec 2022
Apple might have its reservations about the metaverse, but it’s pretty convinced AI art generation will be a part of the future. The tech giant has rolled out beta updates to help ensure a speedy and seamless experience for Stable Diffusion artwork creation across devices.
You’ve seen these text-to-image tools around. All it takes is a text prompt for artificial intelligence to conjure up wholly new artworks according to your vision. The results are often surreal and they’re ready in seconds.
Stable Diffusion may be a relative newcomer next to DALL-E, Google’s Imagen, and Midjourney, but it’s fast gaining popularity with the masses since it’s free and open-source. This automatically opens the gates of AI image creation to the mainstream, and Apple is well aware of that.
As outlined by an Apple machine learning engineer by the name of Atila on Twitter, Apple has boosted Stable Diffusion support on macOS Ventura 13.1 beta four, as well as on betas for iOS 16 and iPadOS. The optimizations ramp up efficiency for Stable Diffusion prompts run on the Apple Neural Engine and Apple Silicon GPU.
Today's release of macOS Ventura 13.1 Beta 4 and iOS and iPadOS 16.2 Beta 4 include optimizations that let Stable Diffusion run with improved efficiency on the Apple Neural Engine as well as on Apple Silicon GPU
— Atila @NeurIPS (@atiorh) December 1, 2022
The machine learning team says an M2 MacBook Air with 8 GB RAM will be able to complete 50 Stable Diffusion iterations in 18 seconds. Meanwhile, “distilled” versions of the machine-learning model—which can only churn out up to four results at once—will only need less than a second to generate artwork on an M2-powered device.
“Since its public debut in August 2022, Stable Diffusion has been adopted by a vibrant community of artists, developers and hobbyists alike, enabling the creation of unprecedented visual content with as little as a text prompt,” explain Apple’s machine learning researchers in a newly-published paper.
“Beyond image generation from text prompts, developers are also discovering other creative uses for Stable Diffusion, such as image editing, in-painting, out-painting, super-resolution, style transfer and even color palette generation,” notes the team.
The company adds that, with the growing number of uses for such technology, it’s essential that devices keep up to support those functions capably.
[via AppleInsider and Apple Machine Learning Research, images via Apple Machine Learning Research]