Apple is widely expected to announce the iOS 18 iPhone software update at WWDC 2024 on June 10 and that update is thought to be laden with new Apple AI features. We don't yet know for sure what those features are, but there have been reports of privacy and speed considerations being made.
It's been suggested that Apple will make its AI features faster and more private than those offered elsewhere by keeping the large language models (LLMs) that make them possible on the iPhone itself. With no need to send data to a server and wait for a response, processing times should be improved. And the privacy implications are obvious.
Running LLMs on iPhones instead of cloud servers is more difficult due to battery life, differing processing capabilities, and more. But Apple has taken a step forward by making new LLMs available to the public as an open-source project.
iOS 18 and beyond
The project, called OpenELM, is available on the Hugging Face Hub which is a community for sharing code that is designed for AI purposes. There are multiple LLMs that have been made available, with the aim being for them to run on a user's device rather than in the cloud.
"We introduce OpenELM, a family of Open-source Efficient Language Models," the project's explainer reads. "OpenELM uses a layer-wise scaling strategy to efficiently allocate parameters within each layer of the transformer model, leading to enhanced accuracy. We pretrained OpenELM models using the CoreNet library." The document then goes on to indicate how the project can be used.
This isn't the first time that we've heard of Apple's plans to run AI models on iPhones of course, but this is another indication of the company's direction of travel. There are still a few months before iOS 18 will be previewed and it won't ship to the public until September alongside Apple's next best iPhone, the iPhone 16, leaving plenty of time for Apple's engineers to get to work.
