Re:Apple Intelligence for iPhone, iPad, and Mac
by dylan604 on Monday June 10, 2024 @03:01PM
(#40637137)
The interesting thing I took from that is they are making servers with M-series chips. Maybe they're just rack mounted Mac Minis? But if Apple decides that crap, maybe it'll get them to a point where they decided to make a proper rack mountable form factor available??? <genmojiTechPrayingInFrontOfRack>
Re:Apple Intelligence for iPhone, iPad, and Mac
by Almondsetat on Monday June 10, 2024 @03:14PM
(#40637388)
I think it's a little disingenuous to think on-device accelerators on a mobile phone would be able to do literally any AI task without help
Re:Apple Intelligence for iPhone, iPad, and Mac
by ducadveritatem on Tuesday June 11, 2024 @06:45PM
(#40652528)
Based on their presentations yesterday and accompanying written materials they have a pretty capable ~3B parameter Foundational LLM running fully locally on the devices which is the first line. https://machinelearning.apple.com/research/introducing-apple...
Re:Apple Intelligence for iPhone, iPad, and Mac
by onesociety2022 on Monday June 10, 2024 @05:23PM
(#40639297)
Maybe that is actually good for iPhone buyers? Otherwise every 2 years, they will claim a bunch of new AI features will not work on your older device. But if they can delegate requests to a server, older devices can continue receiving newer AI features from future iOS releases (they will just be slower than the newer iPhones which will run them locally).
If so that's somewhat disappointing given how much AI power Apple hardware packs.
> Apple sets a new standard for privacy in AI, with the ability to flex and scale computational capacity between on-device processing and larger, server-based models that run on dedicated Apple silicon servers