Apple Releases Open Source AI Models That Run On-Device

Apple today released several open source large language models (LLMs) that are designed to run on-device rather than through cloud servers. Called OpenELM (Open-source Efficient Language Models), the LLMs are available on the Hugging Face Hub, a community for sharing AI code.

Apple Silicon AI Optimized Feature Siri
As outlined in a white paper [PDF], there are eight total OpenELM models, four of which were pre-trained using the CoreNet library, and four instruction tuned models. Apple uses a layer-wise scaling strategy that is aimed at improving accuracy and efficiency.

Apple provided code, training logs, and multiple versions rather than just the final trained model, and the researchers behind the project hope that it will lead to faster progress and "more trustworthy results" in the natural language AI field.

OpenELM, a state-of-the-art open language model. OpenELM uses a layer-wise scaling strategy to efficiently allocate parameters within each layer of the transformer model, leading to enhanced accuracy. For example, with a parameter budget of approximately one billion parameters, OpenELM exhibits a 2.36% improvement in accuracy compared to OLMo while requiring 2x fewer pre-training tokens.

Diverging from prior practices that only provide model weights and inference code, and pre-train on private datasets, our release includes the complete framework for training and evaluation of the language model on publicly available datasets, including training logs, multiple checkpoints, and pre-training configurations.

Apple says that it is releasing the OpenELM models to "empower and enrich the open research community" with state-of-the-art language models. Sharing open source models gives researchers a way to investigate risks and data and model biases. Developers and companies are able to use the models as-is or make modifications.

The open sharing of information has become an important tool for Apple to recruit top engineers, scientists, and experts because it provides opportunities for research papers that would not normally have been able to be published under Apple's secretive policies.

Apple has not yet brought these kinds of AI capabilities to its devices, but iOS 18 is expected to include a number of new AI features, and rumors suggest that Apple is planning to run its large language models on-device for privacy purposes.

Popular Stories

Apple Intelligence General Feature

Apple Intelligence Features Not Coming to European Union at Launch Due to DMA

Friday June 21, 2024 9:44 am PDT by
Apple today said that European customers will not get access to the Apple Intelligence, iPhone Mirroring, and SharePlay Screen Sharing features that are coming to the iPhone, iPad, and Mac this September due to regulatory issues related to the Digital Markets Act. In a statement to Financial Times, Apple said that there will be a delay as it works to figure out how to make the new...
Apple WWDC24 Apple Intelligence hero 240610

Apple Explains iPhone 15 Pro Requirement for Apple Intelligence

Wednesday June 19, 2024 4:48 am PDT by
With iOS 18, iPadOS 18, and macOS Sequoia, Apple is introducing a new personalized AI experience called Apple Intelligence that uses on-device, generative large-language models to enhance the user experience across iPhone, iPad, and Mac. These new AI features require Apple's latest iPhone 15 Pro and iPhone 15 Pro Max models to work, while only Macs and iPads with M1 or later chips will...
amazon echo dot

Amazon Could Charge Up to $10/Month for Alexa

Friday June 21, 2024 2:55 pm PDT by
Apple competitor Amazon is working on a revamp of its Alexa assistant, and the new version could cost up to $10 per month, according to a report from Reuters. The upcoming version of Alexa will support conversational generative AI, and Amazon is planning for two tiers of service. There will be a free tier and a second, premium tier that is priced at $5 at a minimum, with Amazon considering...
General Spotify Feature

Spotify Launches Cheaper $10.99/Month Premium Plan Without Audiobooks

Friday June 21, 2024 4:22 pm PDT by
Spotify today announced the launch of a new Basic paid plan that offers a small discount for dropping access to audiobooks. Priced at $10.99 per month, the Basic option includes all of the music benefits of Spotify like ad-free playback, but without added monthly audiobook listening time. The $11.99 standard Premium Individual plan that Spotify offers includes ad-free playback and 15 hours...
iPhone 16 Pro Max Generic Feature 2

5 Biggest Changes Rumored for iPhone 16 Pro Max

Wednesday June 19, 2024 5:00 am PDT by
Given Apple's rumored plan to add an all-new high-end tier to its iPhone 17 series in 2025, this could be the year for Apple to bring its boldest "Pro Max" model to the table — the kind of iPhone 16 upgrade that stands tall above its siblings, both figuratively and literally. If you have been holding out for the iPhone 16 Pro Max, here are five of the biggest changes rumored to be coming...

Top Rated Comments

Populus Avatar
9 weeks ago

What? No comments? Hey, ChatGPT, generate me a comment....
[Hey ChatGPT, please generate a comment in the style of a typical MacRumors average user, with a touch of acid humor, regarding this piece of news.]

"Oh great, Apple's finally joining the open-source party—just a decade late and probably still with strings attached somewhere in those 'open' terms. They're throwing us a bone with OpenELM, but let's be real, they’re probably just doing it to lure in some AI hotshots tired of their corporate overlords. Now we just have to sit back and wait for iOS 18, where they'll inevitably limit these models to the latest hardware, forcing us all to upgrade. Because, you know, my current iPhone can't possibly handle a couple more AI tricks without combusting."

Attachment Image
Score: 78 Votes (Like | Disagree)
NT1440 Avatar
9 weeks ago

It's clear they are behind and need help. OpenAI's GPT-4 has 1.76 trillion parameters. Apple's OpemELM model has 3 billions parameters.
Does ChatGPT even attempt to run on device? You know, the whole point of this?

The thing I’ve noticed about all these AI hype-people is they certainly know what the “leader” of the pack day to day is, but somehow can’t imagine smaller models being a better solution for a given task. Instead of a one size fits all approach, what’s wrong with invoking a trained AI that is smaller but specifically suited for the task at hand *automatically based on the context of what you’re currently doing*?

Just some food for thought…
Score: 42 Votes (Like | Disagree)
heretiq Avatar
9 weeks ago

Does ChatGPT even attempt to run on device? You know, the whole point of this?

The thing I’ve noticed about all these AI hype-people is they certainly know what the “leader” of the pack day to day is, but somehow can’t imagine smaller models being a better solution for a given task. Instead of a one size fits all approach, what’s wrong with invoking a trained AI that is smaller but specifically suited for the task at hand *automatically based on the context of what you’re currently doing*?

Just some food for thought…
Thank you for adding something meaningful to this otherwise nauseatingly cynical thread. ??
Score: 22 Votes (Like | Disagree)
socialwill Avatar
9 weeks ago

All this for an improved Siri is crazy ngl
Yep, it will now reply 10x faster with "I'm sorry, I seem to be having a problem right now"
Score: 16 Votes (Like | Disagree)
Arislan Avatar
9 weeks ago

It's clear they are behind and need help. OpenAI's GPT-4 has 1.76 trillion parameters. Apple's OpemELM model has 3 billions parameters.
And does that 1.76 trillion parameter model run on a phone without cloud servers?
Score: 16 Votes (Like | Disagree)
heretiq Avatar
9 weeks ago

Apple finally decides to join the open-source party, huh? Only a decade late and I bet there are still some hidden strings attached to their so-called 'open' terms. They're throwing us a bone with OpenELM, but let's be real, they're probably just trying to lure in some AI hotshots tired of their corporate overlords. And now we have to wait for iOS 18, where they'll undoubtedly limit these models to the latest hardware, forcing us all to upgrade. Because, you know, my current iPhone surely can't handle a few more AI tricks without spontaneously combusting. Classic Apple move.
Hmmm .. FYI: it seems that @Populus beat you to using ChatGPT to “contribute” to the discussion; only difference is that he acknowledged he was using it .. ?
Score: 15 Votes (Like | Disagree)