Hopefully Apple building its own AI servers mean they're not about to start relying on Google Gemini of all things for AI. 😬
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Reportedly Building M2 Ultra and M4-Powered AI Servers
- Thread starter MacRumors
- Start date
- Sort by reaction score
Training models like LLMs requires huge resources. On-device there will only be the finished model.
I was hoping it would be on device processing. That's a shame. So they better have a good excuse for reserving certain AI features for iPhone 16 Pro.
the bigger question: how will this benefit YOU?Hopefully this will benefit macOS
The whole narrative went quickly from "new leak shows all kinds of AI features in iOS 18 OMG" to "the real AI features will come to iPhone 17 (2025), almost 3 years(!) after the launch of ChatGPT".
My god, I will be listening to WWDC this year but if they don't play ball quickly and show that they are a development powerhouse, I might leave the whole ecosystem in a year of two. The costs are staggering (1200 bucks for a so so phone) and even for my hobby in AI I had to resort back to Windows (with WSL). I came to this "camp" with the introduction of M1 (and I am typing this on that very MBP which is great for everything not-machine learning). I was about to switch my iPhone 12 Pro to a iPhone 16 Pro, but if they don't come up quickly with things that would actually make my live easier, I will reconsider. Spending 1200 EUR on something that is feature-wise on par with an 2021 Android is just.... unbelievable.
Not to mention their malicious compliance with EU regulation. While typing this, I am more and more thinking "what the hell am I still doing with Apple". I notice it on the job as well; disappointment and apple is becoming the joke.
My god, I will be listening to WWDC this year but if they don't play ball quickly and show that they are a development powerhouse, I might leave the whole ecosystem in a year of two. The costs are staggering (1200 bucks for a so so phone) and even for my hobby in AI I had to resort back to Windows (with WSL). I came to this "camp" with the introduction of M1 (and I am typing this on that very MBP which is great for everything not-machine learning). I was about to switch my iPhone 12 Pro to a iPhone 16 Pro, but if they don't come up quickly with things that would actually make my live easier, I will reconsider. Spending 1200 EUR on something that is feature-wise on par with an 2021 Android is just.... unbelievable.
Not to mention their malicious compliance with EU regulation. While typing this, I am more and more thinking "what the hell am I still doing with Apple". I notice it on the job as well; disappointment and apple is becoming the joke.
GPT stands for Giant People Tracker.
Bet you didn’t know that.
Bet you didn’t know that.
Is that joke alluding to the issue that Microsoft uses Linux on their servers?
Im all for this… their SoC are massively powerful, and with their own IP and software…. Never underestimate the power of vertical integration.
Hear me out:
What if Apple made a consumer server that served as the backend for a “local” GenSiri? Something that your iProducts could call back to for AI interactions maintaining privacy.
Probably prohibitively expensive. Nevermind. I’ll go back to hoping for a new AirPort Extreme mesh router.
What if Apple made a consumer server that served as the backend for a “local” GenSiri? Something that your iProducts could call back to for AI interactions maintaining privacy.
Probably prohibitively expensive. Nevermind. I’ll go back to hoping for a new AirPort Extreme mesh router.
No, telling the fact cause Apple Silicon's GPU performance sucks. That's why Mac is only good at 2D based software.
I had the feeling that this "on device" propaganda was not going to be what you would expect. What is called "AI" these days is basically big data rather than real AI, and big data means that datasets take a huge space. Does "on device" mean that you will have your huge datasets on your device? I doubt that's Apple's plan, they'd rather prefer to control the datasets. In the end, "on device" will be just an ad slogan, but I guess the real implementation will be a disappointing cloud-based system.
How does this jive with the previous reports that Apple was making a custom AI chip? Will that chip just be an M4, or an M4-variant? Or a truly specialized chip? Sounds like they are just building an AI server-farm using existing Apple Silicon. Is that a stop-gap until they have a new chip?
This move is ironic. While at Apple, the founder of Nuvia wanted to design ARM-based server processors but ended up leaving because Apple, at that time, didn't want to. And now Apple does. They could have kept that talent in-house.
With OpenAI, Microsoft, Amazon, Tesla, and everyone else building their own AI server chips and lamenting the supply shortage of server chips plus being at the mercy of Nvidia, it makes a lot of sense. It makes so much sense that they should have been making them a long time ago and Johny Srouji should quit saying in every interview that a computer manufacturing company that builds and designs its own chips for ALL its products, is not in the computer chip business.
hahahaha for some people like you hate is a drive. It moves you up here to troll on this forum. But you are not well informed.
M3 family solved the 3D performance. It is the most powerful processor family in terms of performance per watt.
And if you want to see where all of this is leading to. Just wait until WWDC for the new Mac Studio and Mac Pro.
maybe that hate is just envy.
M3 is far from being good with graphic card performance. Beside, where are those 3D software to work with? Even M3 Max is barely better than mobile RTX 4060 and it doesn't even have options for high-end, especially RTX 4090 and yet you are claiming it's great. Be aware that Mac is only well known for 2D based such as FCPX and Logic Pro. Also, M3 series are 3nm based while RTX 40 series are 5nm which is two generations ahead. M2 series which also used 5nm aren't even close to RTX 40 series and M3 series are still even close to it. Performance by watt only proves that M3 series sucks when the performance matter.
It's so funny whenever people claim that Apple Silicon's GPU performance is great which is clueless. If you hate to admit the fact, why not research more on the actual fact first?
I can guarantee you they'll be ordering from AMD the MI300+ series. These AI Servers with their own models won't be hitting their datacenter systems for at least 18 months and even then those won't be the only option installed.
I don’t know what you are talking about. The LLMs are not big data unless you call lots of parameters data..
100% agreed. How can they be so slow at catching up?
You're one of those Apple is doomed people!
I get it but Apple is not doomed! First to market is not the Apple Way! Perfecting the market is Apple's way! There were smart phones before the iPhone and tablets before the iPad and all of them do not exist now!