3 ms·
It's going to be an interesting conundrum for Apple, with their drive to keep everything on phone. Both the computing requirements and the data requirement on
by whartung 3y ago
It's going to be an interesting conundrum for Apple, with their drive to keep everything on phone.
Both the computing requirements and the data requirement on modern AI make it a challenging fit. So it will be a curiosity about how Apple tries to balance a feature set that they can power on the phone, net take advantage of the modern AI capabilities.
Or, just bail, and funnel it to the net.
- wallaBBB 3y agoMy initial reaction was completely opposite to yours: - Oh this is going to burn google's money, especially if the phone becomes popular. Unless you want to charge a subscription for a feature it really isn't smart for Apple to try something like this.
- foota 3y agoAren't the M1 and friends relatively capable for LLM tasks?
- brucethemoose2 3y agoSort of. Apple is super stingy about RAM capacity on their devices, which is a problem for GenAI Also, the M1 NPU (the equivalent to Google's offloading) is not very fast in Apple's own Stable Diffusion port, and I'm not aware of any community LLM frameworks that even use the NPU. They all run Metal implementations on the M1's GPU.
- ActorNightly 3y agoApple doesn't care about on or off phone, its only goal is to make trendy devices that sell because of the badge.
- astrange 3y agoApple spends over $20 billion a year on R&D.
- ActorNightly 3y agoWhats your point? Their device features has been behind android for quite some time, with only a slight edge in overall maxed out performance.