MacBook Pro with M5 Pro and M5 Max • Hacker News Board

Re: MacBook Pro with M5 Pro and M5 Max

#141

Post by SirMaster » Tue, Mar 03, 2026, 2:53 PM UTC

jmyeet wrote: post #106

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.…

Apple absolutely has a massive opportunity here because they used a shared memory architecture. So as most people in or adjacent to the AI space know, NVidia gatekeeps their best GPUs with the most memory by making them eye-wateringly expensive. It's a form of market segmentation. So consumer GPUs top out at 16GB (5090 currently) while the best AI GPUs (H200?) is 141GB (I just had to search)? I think the previou sgen…

>So consumer GPUs top out at 16GB (5090 currently)

5090 has 32GB, and the 4090 and 3090 both have 24GB.

Re: MacBook Pro with M5 Pro and M5 Max

#142

Post by Someone1234 » Tue, Mar 03, 2026, 2:53 PM UTC

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.

I strongly agree. People see local "GPT-4 level" responses, and get excited, which I totally get. But how quickly is the fall-off as the context size grows? Because if it cannot hold and reference a single source-code file in its context, the efficiency will absolutely crater.

That's actually the biggest growth area in LLMs, it is no longer about smart, it is about context windows (usable ones, note spec-sheet hypotheticals). Smart enough is mostly solved, combating larger problems is slowly improving with every major release (but there is no ceiling).

Re: MacBook Pro with M5 Pro and M5 Max

#143

Post by m3kw9 » Tue, Mar 03, 2026, 2:53 PM UTC

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.…

A useful llm that needs 64gb of ram and mid double digit cores is not useful for 99% of their customers. The LLMs they have on iphone 17's certainly cannot do anything useful other than summerization and stuff. It's a hardware constraint that they have.

Re: MacBook Pro with M5 Pro and M5 Max

#144

Post by freeone3000 » Tue, Mar 03, 2026, 2:53 PM UTC

Hamuko wrote: post #67

butILoveLife wrote: post #45

Earlier quoted context omitted.

I think its just marketing, and the marketing is working. Look how many people bought Minis and ended up just paying for API calls anyway. (Saw it IRL 2x, see it on reddit openclaw daily) I don't mind it, I open Apple stock. But I'm def not buying into their rebranding of integrated GPU under the guise of Unified Memory.

I've tried to use a local LLM on an M4 Pro machine and it's quite painful. Not surprised that people into LLMs would pay for tokens instead of trying to force their poor MacBooks to do it.

I’m super happy with it for embedding, image recog, and semantic video segmentation tasks.

Re: MacBook Pro with M5 Pro and M5 Max

#145

Post by davidmurdoch » Tue, Mar 03, 2026, 2:54 PM UTC

AdamN wrote: post #127

davidmurdoch wrote: post #75

Earlier quoted context omitted.

Who is "us" in this case? Majority of devs that took the stack overflow survey use Windows: https://survey.stackoverflow.co/2025/technology/#1-computer-...

That's the broad developer community. 90%+ of the engineers at Big Tech and the technorati startups are on MacOS with 5% on Linux and the other 5% on Windows.

Source?

Re: MacBook Pro with M5 Pro and M5 Max

#146

Post by macintux » Tue, Mar 03, 2026, 2:54 PM UTC

butILoveLife wrote: post #53

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.

[flagged]

Please read the guidelines and consider moderating your tone. Hostility towards other commenters is strongly discouraged.

Re: MacBook Pro with M5 Pro and M5 Max

#147

Post by tamimio » Tue, Mar 03, 2026, 2:54 PM UTC

I will wait for the new mac mini instead

Re: MacBook Pro with M5 Pro and M5 Max

#148

Post by vile_wretch » Tue, Mar 03, 2026, 2:54 PM UTC

hu3 wrote: post #74

ezfe wrote: post #47

Earlier quoted context omitted.

They’re literally not changing

It did change. They bumped $200 on the entire line. So even the 16GB version is more expensive. I'd love to have customers like Apple. Bumps $200: "it didn't change!!!" And no power adapter included.

The EU forbids them from including power adapters. They're still included everywhere else.

Re: MacBook Pro with M5 Pro and M5 Max

#149

Post by freeone3000 » Tue, Mar 03, 2026, 2:54 PM UTC

whizzter wrote: post #26

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.…

We had a workshop 6 months ago and while I've always been sceptical of OpenAI,etc's silly AGI/ASI claims, the investments have shown the way to a lot of new technology and has opened up a genie that won't be put back into the bottle. Now extrapolating in line with how Sun servers around year 2000 cost a fortune and can be emulated by a 5$ VPS today, Apple is seeing that they can maybe grab the local LLM workloads if…

Torch mlp support on my local macbook outperforms CUDA T4 on Colab.

Re: MacBook Pro with M5 Pro and M5 Max

#150

Post by satvikpendem » Tue, Mar 03, 2026, 2:55 PM UTC

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.

That should be covered by the harness rather than the LLM itself, no? Compaction and summarization should be able to allow the LLM to still run smoothly even on large contexts.