Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

MacBook Pro with M5 Pro and M5 Max

apple.com

Post reply on HN
Print view View original

501–510 of 1001 posts

Re: MacBook Pro with M5 Pro and M5 Max

#501

Post by travisjungroth »

brailsafe wrote: post #492

I don't see it mentioned much, but the most exciting thing to me is that they're shipping their own WiFi chip in it, which leads me to be hopeful that they'll eventually get around to shipping a cell modem so I don't have to tether to my phone constantly. Still no new colours unfortunately. I think those are the two things that would/will be exciting in the future. Give me a green 5g+ capable MBP and I'll be happy. I…

I don’t know if they’ll ever do that. Colors add another dimension, so you either need to have more stock on hand or do more custom models. Right now, the profit margins on all upgrades is huge.

Phones have less configurability, they sell more, and colors seem more important.

Re: MacBook Pro with M5 Pro and M5 Max

#502

Post by iAMkenough »

game_the0ry wrote: post #65
Tangokat wrote: post #5

"Scaling up performance from M5 and offering the same breakthrough GPU architecture with a Neural Accelerator in each core, M5 Pro and M5 Max deliver up to 4x faster LLM prompt processing than M4 Pro and M4 Max, and up to 8x AI image generation than M1 Pro and M1 Max." Are they doubling down on local LLMs then? I still think Apple has a huge opportunity in privacy first LLMs but so far I'm not seeing much execution.…

> Are they doubling down on local LLMs then? Honestly, I think that's the move for apple. They do not seem to have any interest in creating a frontier lab/model -- why would they give the capex and how far behind they are. But open source models (Kimi, Deepseek, Qwen) are getting better and better, and apple makes excellent hardware for local LLMs. How appealing would it be to have your own LLM that knows all your se…

RE Frontier models/hardware: I'm interested to see what happens with their "private cloud compute" marketing concept now that they're moving from running Siri AI experiences on Apple servers to Google servers instead.

Re: MacBook Pro with M5 Pro and M5 Max

#503

Post by pmdr »

prodigycorp wrote: post #49
testfrequency wrote: post #15

I have a fairly maxed out M2 Ultra (24 cores, 192GB RAM), and still cannot get this machine to choke on anything. I have not once felt the need to upgrade in years, and that’s with doing pretty demanding 3D and LLM work.

If there’s anything this past three years has taught me, it’s that modern cpus can performantly do every task except for streaming text over the internet.

I had to upgrade the CPU in a 10-year old machine (from i5 to i7) to have decently -working javascript on websites. Every other piece of software worked fine, though.

Re: MacBook Pro with M5 Pro and M5 Max

#504

Post by zozbot234 »

hu3 wrote: post #81
dirk94018 wrote: post #37

On M4 Max 128GB we're seeing ~100 tok/s generation on a 30B parameter model in our from scratch inference engine. Very curious what the "4x faster LLM prompt processing" translates to in practice. Smallish, local 30B-70B inference is genuinely usable territory for real dev workflows, not just demos. Will require staying plugged in though.

What about real workloads? Because as context gets larger, these local LLMs aproxiate the useless end of the spectrum with regards to t/s.

The thing about context/KV cache is that you can swap it out efficiently, which you can't with the activations because they're rewritten for every token. It will slow down as context grows (decode is often compute-limited when context is large) but it will run.

Re: MacBook Pro with M5 Pro and M5 Max

#506

Post by efxhoy »

giancarlostoro wrote: post #437
lastdong wrote: post #423

Earlier quoted context omitted.

14-billion parameter model with 4-bit quantization seems rather small

On my 24GB RAM M4 Pro MBP some models run very quickly through LM Studio to Zed, I was able to ask it to write some code. Course my fan starts spinning off like the worlds ending, but its still impressive what I can do 100% locally. I can't imagine on a more serious setup like the Mac Studio.

How is the output quality of the smaller models?

Re: MacBook Pro with M5 Pro and M5 Max

#507

Post by robotresearcher »

Greed wrote: post #430
ttoinou wrote: post #275

Earlier quoted context omitted.

With M3 Max with 64GB of unified ram you can code with a local LLM, so the bar is much lower

But why? Spending several thousand dollars to run sub-par models when the break-even point could still be years away seems bizarre for any real usecase where your goal is productivity over novelty. Anyone who has used Codex or Opus can attest that the difference between those and a locally available model like Qwen or Codestral is night and day. To be clear, I totally get the idea of running local LLMs for toy reason…

Sometimes you can't push your working data to third party service, by law, by contract, or by preference.

Re: MacBook Pro with M5 Pro and M5 Max

#508

Post by astrange »

wincy wrote: post #223
detritus wrote: post #212

Earlier quoted context omitted.

Don't these integrated ARM-based SoCs make much better use of RAM as opposed to old Intel-based boards? That's my understanding, anyway.

My wife’s 8GB MacBook Air crashed yesterday with Firefox and Find My open and nothing else because of running out of RAM, so, sort of, but they’re not magic. (Find My was using 3GB of memory!)

If you mean it showed the out of memory dialog, that wouldn't be caused by an app using 3GB. The dialog shows up at ~48GB swap space used on an 8GB Mac, or when you're out of disk space and can't write a swap file.

Re: MacBook Pro with M5 Pro and M5 Max

#509

Post by astrange »

wincy wrote: post #310
nozzlegear wrote: post #294

Earlier quoted context omitted.

So Firefox was using 5gb? There's your problem.

It’s a losing battle me trying to tell my wife to close her Firefox tabs, haha, but yes, Firefox does use a lot of ram when you have 500 tabs. Maybe I’ll get her a 64GB MacBook Pro for the premium web browsing experience she so desires!

She shouldn't have to do anything. That's the browser's job.

Re: MacBook Pro with M5 Pro and M5 Max

#510

Post by myHNAccount123 »

dmix wrote: post #236
heurs wrote: post #131

Honest question. Is it possible to install an earlier version of macOS on these machines? Liquid glass looks so.. unprofessional to my eyes. And I hear it's also unstable.

You barely see any liquid glass on Tahoe. I keep my dock hidden and it's just the icons mostly which aren't that different than before.

Same here. Not really understanding the complaints for macOS. I think the addition of icons in the context and menus is worse than glass.
Post reply on HN