412 comments

[ 0.22 ms ] story [ 10.0 ms ] thread
Mac Studio with M5 Max starts at $2,499 (U.S.) and $2,299 (U.S.) for education. Additional configure-to-order options are available at apple.com/mac-studio. Mac Studio with M5 Ultra starts at $5,499 (U.S.) and $5,099 (U.S.).

I am in Europe, and the Mac Studio M5 Ultra GPU 64 cores with 96GB RAM is up to 6.649,00 €. Ouch.

Don't forget that US prices usually do not include the VAT, while EU prices usually do include respective VAT.
A lot of us in America have no sales tax (if that’s what you mean by VAT) and those that do have it at a fraction of EU VAT rates.
> A lot of us in America have no sales tax

Not that many of us, actually. Only Montana, New Hampshire, Oregon, and some parts of Delaware and Alaska have no sales tax. https://commons.wikimedia.org/wiki/File:Sales_tax_by_county....

(comment deleted)
(comment deleted)
To put this into a perspective, Google helpfully reminds:

> A fully configured IBM Personal Computer AT (Model 5170) with expanded memory and storage cost around $5,795 to $6,000 at its launch in August 1984, which equals roughly $18,600 to $19,300 in 2026 USD.

It would be interesting to compare the costs of a top of the line machine every decade or so. Costs were steadily decreasing until recently.
They still are, if what you want is roughly the same as the prior generations capability with some uplift (making then number up, but say 20% faster or more ram or whatever).

What is changing is that there genuine demand for more capabilities disproportionate to the cost decrease curve. Fab demand and supply constraints have slowed or even reversed some cost decreases - but that is still getting absorbed by the overall systems costs when you are looking at things like laptops. If you all you want is the last decades demand to browse the web and use office - things are cheaper than ever.

John Dvorak said many, many decades ago (80s/90s) that the computer you want will always cost $3000. That statement has been more/less true for some time periods than others, but with some wiggle room I’ve found it to be accurate enough.

Care to guess the approximate price of the MBP I bought earlier this year?

I bought my refurbished M3 Max MBP with 64 GB for $3k a couple of years ago. Before Ethan I never spent more than $2k for a computer.
The “spend” and “want” number might differ, depending on one’s financial state. I know I’ve purchased plenty for less than $3K. But the one I wanted
I'm old and established (after spending most of my life not), but ya, I get what you are saying. It was a huge leap for me to spend $3k on a laptop.

Before local LLMs became a thing, I had completely lost interest in buying anything but the cheapest laptop. It felt like "personal computing" was a solved problem. But...it actually isn't, and this is exciting.

My heavily-upgraded M1 Max came in slightly over that when I got it five years ago. (Still going very, very strong.)

This new Studio? Can't find a config under $5k I'd bother with. But for the MBPs that number still mostly tracks for the average Pro user. (I buy large and run it into the ground so long I mistake the ground for the computer's remains.)

I got a 64GB M1 Max four years ago or so, used. Felt like a ridiculous overspend at the time, but I'm sure smug about it now.
> I am in Europe, and the Mac Studio M5 Ultra GPU 64 cores with 96GB RAM is up to 6.649,00 €. Ouch.

It would be significantly cheaper to fly to a tariff-free country and buy there.

Technically, California and other state have laws that say you must report this purchase and pay the tax the moment it enters the state.
Bizarre there isn’t a 1TB RAM option hidden away for the excessively frivolous or VC funded.
Same reason they cut the big options on the existing models, this way they can sell more devices. The additional cost for the additional 512GB would have to make up for the loss of another sold device otherwise. No idea if there would really be that many people buying this then while on the other hand AI stuff makes people do crazy stuff, so...yeah :)
Considering Apple pricing it actually might.

256GB model is $10k and the 512GB version will probably be double

They seem to be suffering from the supply constraints like everyone else. They phased out the higher capacities on the M3 Ultra Mac Studio a while ago, and if you order a 128GB MBP, say, you're looking at six weeks or more for delivery.
It's worse than six weeks: Apple quoted me 10-12 weeks yesterday for an M4 Max Studio (about Nov 17) December for the M3 Ultra. Planning to lock in a new model soon. I went looking for used but those are wildly expensive. Strange times. https://hard.bargains/posts/mac-before-sept-22/
It's more bizarre that Apple got caught with pants down. Focused on CPUs and ignored RAM.

Seems like miscalculation. If they had their own fab for RAM, they could completely corner the market today.

They have no fab for CPUs, they are manufactured by Samsung and TSMC. The bottleneck is in manufacturing RAM not CPUs so there is nothing Apple can do here.
Let me rephrase. They have not booked enough capacity at contract manufacturer.
Recently Micron pointed that finger and threw that shade in Apple’s direction saying that their priority for years has been squeezing margins in their favour, so there wasn’t the slack in the system to build the fab capacity that is now desperately required given the sudden need for RAM now that there’s a cause to use it in AI Inference.
Apple has a fab for CPU R&D. It's not for final production though.
They don’t have their own fab for CPUs either.
They would have had to start building that fab 5-10 years ago. It would have been incredible foresight to do so and it would have looked insane.
512GB unified memory option coming available in October.
The 256GB option is +$4,000 - the overall price for 512GB setup would probably be $20k!
It needs to be a little less than double the overall price for 256GB option, otherwise it's better to get 2 256GB and link them.
you can't link onchip memory.
True, but you can link them up over thunderbolt or Ethernet. If your goal is to run local LLMs, not all weights need to live on the same computer. You can segment the workload by layers and pass the activations along the lower bandwidth interconnect with a small perf penalty. Also you get double the CPU/GPU cores allowing for better multi user/agent performance.
Ah sorry you are right I missread what you meant by it.
I was thinking the same as you as far as price per value, it does make seance to get 2x of these things IMO, but what throughput hit would you see in linking versus one machine? latency does matter, and there must be a trade off no?
512GB unified RAM is going to be really good for running local LLMs.
What is unified ram? RAM and VRAM as one thing?
In best case its also on-chip high speed ram.
> M6 supports up to 32GB of unified memory to multitask across demanding apps and run LLMs on device for secure and private agentic tasks. It also provides up to 170GB/s of unified memory bandwidth — a 10 percent increase over M5 and a 2.5x increase over M1.

Isn't 170GB/s slow for bandwidth?

It is. That's the mac mini. For local LLMs you would want the Mac Studio, which tops out at 1.2TB/s.
Kinda. Strix Halo does 256GB/s of memory bandwidth, and is significantly slower than an M5 Max (614GB/s). Feels like intentional market segmentation?
For max memory bandwidth you need to buy the Ultra versions (M5 Ultra: 1,2TB/s, this gets comparable to real GPUs regarding the memory bandwidth).
It's higher bandwidth than any dual-channel DDR5 desktop machine, but Apple never quote the memory latency, so hard to compare otherwise.
Would that be a difficult comparison to do fairly sine one is SoC and the other isn’t?
I meant that it's a difficult comparison between platforms without having all the details like memory latency, how the memory controller load increases latency (can increase by 4x on some platforms), cache latency, TLB entries, etc.

Extremely high bandwidth is great for copying data, but not as relevant for walking chains of pointers, where latency/cache/TLB entries are more important.

So just saying "high bandwidth == better" is true when other variables are the same, but they rarely are, especially in comparison to x86-64 offerings.

All of these are independent that it is a SoC with on-package DRAM.

You would want to get the M5 pro version with 307gb/s if you were interested in running local LLMs.
M6 and M5 Ultra both have comparable memory bandwidth per GPU core. I think they will perform well.
Due to pricing insanity (not that Apple prices weren’t insane before the ram/ssd shortages) I’m not in the market but I do wonder if my next computer should be a Mac Studio instead of a MBP that lives its life docked. Might be better to just run a Studio and Neo for the very few times I actually need remote capabilities.
I had the same thought, I grabbed a studio two years ago for this reason and it’s been great. 99% of the time lack of portability isn’t a concern. Every now and then (e.g. travel) I notice the limitation, but it’s not much of an inconvenience to just not do some work for a bit.
Plus remote work is getting easier and easier. There are so few instances when I'm not able to get online. If we lived in a world where hardware were getting cheaper, it might make sense to splurge. In this environment I think the Neo is perfect.
Build quality of the Neo is extremely good, I love the keyboard — it’s more tactile and reminds me of early 2010s MacBooks.

I’ll be selling my M4 MBA soon, I genuinely use the Neo more. Huge difference in typing experience.

Great repairability is a plus. It was super easy, and actually fun to open. Felt like unboxing an Apple product. Applied the thermal paste mod for $10 which works excellently; I’ve had it shortly after launch.

And I love the notchless display, even if I wished the color gamut was a bit better.

The Neo keyboard just feels nicer than even an M5 Air.
That’s what I have been doing for years, it remains in the house secured while I ssh into it from an old thinkpad. You can get air to pair it with it if you really wanna have that seamless flow, otherwise, ssh works well.
I was in the same situation, I used maxed out 15'' M3 Max MacBook Pro docked to Studio Display closed on vertical stand behind the screen. It was fine for office work, but running local LLMs would definitely overheat it. The battery started degrading purely due to heat issues. And it was audible as well.

I decided to get Mac Studio M4 Max, also all maxed out config and the cooling is so much better that I can run local LLMs like Gemma 3/4, gpt-oss 120b all day long without any heat issues or any audible fan noise. So for my use case it was the right decision. I subsequently added 15'' M5 Max MacBook Pro all maxed out to my collection and even though it is slightly faster on LLM inference (I get 100 tokens/s with Gemma 4 27b model), you just can't run LLMs longer than a few minutes. It starts overheating and gets really loud.

Weird, I’ve run LLM batch sessions for hours on my Max M3 MBP. It doesn’t get very loud, though I’m not getting anything close to 100 tok/s on a 27b model, I use a 35b MoE model just to get 90 tok/s. The fan comes on but thermally it never overheats. I do have it in a vertical closed position, though.
I've been holding out, because I think my next purchase will be a Studio with an Ultra Chip in it. I'm wanting it to be a "forever" server, so I'm holding out while I can.
If running LLMs locally matters, it’s hard to imagine a “forever machine” existing in anything less than 5-10 years, probably more. This stuff is just evolving so rapidly. Buying a “forever machine” today might be like buying a “forever GPU” in 2003.
How about business where you send in all your old devices and get back a SSD using using their memories
Laptops can't do agentic engineering. They get hot as hell and battery drains instantly. I think this will promote a switch to desktops for the next couple of years, until we have new mobile chips.
Are you referring specifically to agentic engineering with locally hosted models?
I dusted of my lightest computer with an M1 chip and use Tailscale to make my network virtual from anywhere. Been running a. Pi5 as a main house hub and an M1 Pro as an always on Mac. It would be nice to go all out and make a Studio a hub I can just screen share into for major compute.
Do it! I went with the Mini/Neo combo. I don't need MBP power when out and about. When at home, the Mini is all I use.
If all you want to do is remote into your desktop, Neo seems like overkill. Why not just get a $200 Chromebook and save yourself $500?
Because the screen, keyboard, and especially trackpad on a $200 Chromebook sucks?
Also it'll be truly too slow. You need to at least be able to access a shared document while a video call is going.
Because who hates themselves that much? It's the thing I touch and interact with. That's exactly the part that needs to be sturdy, smooth, and pretty. It's the facade to the beast at the other end.
This comment got me curious, so I searched for cheap laptops with mobile internet. There's at least one LTE Chromebook now for about 300 euros. This might be my next laptop if it can run Tailscale and remote desktop.
I've been thinking about this a fair bit recently.

We make a lot of price/performance compromises for having an attached screen and keyboard on our computer. That was what got me started.

Then I remembered the days of having to go to a special corner of the house to use a computer, vs now when I have a computer with me all the time. In my bag, on the sofa, on the train. Hell, I'm writing this on the work MBP while waiting for an appointment.

And you know what, I think I got more done when I went and sat in a corner of the house all those years ago. I set up an area for "computer work", and it worked really well.

I have a home office, but it's a jumble of cables going into docking stations and all sorts of weird stuff. I think if I streamline it and turn it into a proper "computer room", I might get some of that mojo back. I might even convince my partner that surrendering the home office and having a corner of the den might be good - she can watch TV while I tinker. And I won't be balancing a laptop on my knee and trying to do two things at once.

And the price/performance thing comes back in. Hmm.

There is a definite mental aspect for most WFH folks to having a space that is dedicated to work. I'm not unique in saying this, but the way I put it is "If you work from anywhere in your house, then you're always at work."

And that, from mental load standpoint, is not healthy for most folks.

Or, if you're like me then "if you work from anywhere in your house, then you're never at work."
Or you are always at work.
I read it as though he’s “working”, like quiet quitting or something.
Or you're never at work, like they said.
But most accurately, they're always at work.
For most people, I'd reckon they are never at work.

There is a reason we know that remote learning is horrible for most people compared to in-classroom. We proved this decisively during COVID.

There is zero reason to believe that most folks magically change overnight from being incapable of remote learning to being highly capable remote workers. It's just not believable.

I hired remote workers in the 90's. It was a small fraction of the total candidate base that could successfully self-motivate and have the discipline to become high performers in such an environment over the long haul. Most of my interviewing and candidate vetting had to do with the remote aspect vs. technical skillset. Luckily around that time is when open source became a huge thing, so those projects presented a pool of pre-vetted candidates to hire out of. The rest of the candidate pool was a total crapshoot.

Remote working has become easier and the tooling and technology much better. But from where I'm standing - many folks do not take it seriously. Simple stuff like having backup Internet is a filtering question for me even today.

“Simple stuff like having backup Internet” that sounds expensive. Who has room to back up the entire Internet?

Personally on the 0.2 days a year I have to worry about it, I just head to the coffee shop. Or you know … take a couple hours off (shock!).

Also I’ve spent literally decades working with very highly productive people exclusively remotely. None of us find this odd. Not sure why you have that level of suspicion/distrust. Granted, things have changed a lot since the 90s.

Backup imternet? Everyone has a cellphone to tether from these days no?
On some days I wor from home but need to be present and the sun is out. So I take my laptop outside and work from there. I cannot do this with a desktop. Wouldn‘t the workaholic just stay inside and miss out?
Sometimes I'll just pop into a meeting from my phone and walk around. That said , I do have a laptop but only because I go to the office every so often . Otherwise, I work exclusively from my home office because that's where my home desktop computer is and I control everything on my work computer from there.
Camera on or off?

I love taking meetings while walking but since Covid everyone has their camera on making sedentary meetings mandatory.

There's a lot of value in camera-on, but for some meetings, I love being camera off, nowhere near my computer, somewhere else in the house, watering plants or sweeping the floor. Sometimes it actually has me MORE engaged with the verbal aspect of the meeting, listening better, speaking better, because there's less distraction from Slack and Hacker News...
I'm much more attentive in some meetings when my hands aren't on the keyboard. For particularly important meetings I'll get out my knitting so that I can be close by while having something that's not failing unit tests occupying my hands
Off now, but historically even at camera on companies I'd do whatever and just hold my phone when convenient for me or just shove the meeting in my pocket if I need my hands.

Obviously depends on your team culture, but try it :)

Wireless headphones can get you halfway there if they have enough range for ya to walk around the house a bit. Airpods are quite good for this.
What if there was a way to use your home computer from a very cheap laptop?
I've done that, but I chose a Framework Desktop instead. The latest Fedora is closer to Snow Leopard than anything Apple has to offer. Downside is that I still have my MBP because of the lock-in and occasionally pick it up to do computer stuff in weird places.
I got used to the hub-and-spoke model at home (previously thin terminal, server-client, etc.). Big ole desktop/server and smaller devices that (ab)use it remotely. Roam around with a smaller computer/tablet/phone. Tailscale to bind it all together.

If your computing needs line up, it's a very serviceable approach.

Yes, this is my approach now: a home server with devices connected via Tailscale. I'm starting to do more from my phone and less on the laptop.

I haven't added my iPad to the Tailnet yet but i reckon that it could become a very comfortable and productive device for me.

While I do love Fedora, I Wouldn't call it a "Snow Leopard". Around August 5th, there's been numerous regressions when it comes to AMD graphics. First a kernel regression that resulting in artifacts on the desktop. Then a linux-firmware regression with decoding AV1 videos that resulting in 4 prominent columns and constant major color shifts.

Its update policy can result in this sort of regressions in a middle of a release. One thing that's nicer thing about macOS is that a released version doesn't regress as much (even if they are buggier at release; but in that case you can just wait until a later point release to upgrade).

I used Fedora for a long time but ended up switching to Ubuntu. Primarily because Fedora seemed to require a reboot after almost every update.

From memory it was a bit better if you updated from the command line but it was still a pain point.

The two advantages I find that macOS has over both Ubuntu and Fedora is support for proprietary music apps and some of the accessibility features seem better.

For day to day work Linux in general has always been pretty good for me.

BRING BACK THE COMPUTER ROOM
A fathers gift
We're turning your bedroom back into the computer room
The wired network connection is just so good.

But the couch is just so comfy.

To me the idea of working on (non-docked) laptop always seemed like an idea that you would do only if there is no other possible way.

Screen is small and is only one. Ergonomics is entirely messed up. Either your screen is too low, or your keyboard is too high. Keyboards are non-ergonomic and have to be made with compromises due to height limits. Touchpad instead of mouse/trackball is compromise for many - and also stuck at one position.

And yet they have somehow spread despite number of people going on business trips not really increasing.

You can dock your laptop at work, then dock it at home, take it with you to a meeting room at work to browse supplemental materials and/or project your screen n the meeting, and use it for the occasional trip (like using it on the train to work).

It's not like you have to be on a plane for the portability to be useful.

I think the point of OP is that while laptops are fine for business trips, tethered usage is superior. My setup is a laptop. Both at home and as work I have a solid desk with an ergonomic chair. On the desk are thunderbolt docks, the docks have hooked the keyboard I like, 2 monitors on mobile stands. And that's it. I have most the advantage of a desktop while using the same machine. One cable to unplug. Super nifty.
I see it now, I misread.

Still, being able to take you laptop to a meeting to either project, or to look up related material to catch up and clarify before making a point in a meeting etc. is invaluable (no, working at a laptop during a meeting doesn't mean you're not paying attention, it often means you're cross-referencing discussions with other data live).

It sounds like you don't code much in bed or with your feet up.

A laptop is a distinct tool from a desktop and if you try to use it like a desktop, I agree it is a terrible substitute. Personally I find laptops vastly more ergonomic than a desktop, but I never program with my feet on the ground.

> code much in bed

I've found my smartphone's the best device for that. I wrote an entire programming language using my phone.

Been trying to make a handheld cyberdeck to replace the phone with something better, but it's still a long way from being real.

Can you share more about your setup? I sort of hate my laptop. I carry a ridiculously overpowered phone around with me all the time, though, so the dream is just to dock it at home or at work and just use that. It’s a shame that Apple does not make this a better experience with iPhones; I would happily shell out for the most expensive iPhone if it were a real desktop replacement. Being able to develop (for real, with a terminal) is a must though.
Android phone with termux and all the usual Linux tools. I use vim to write code, and I type with the touch screen. Not exactly peak ergonomics, but it works. For me, ability to code while lying down comfortably beats having a physical keyboard and bigger screen.

Set up wireguard in my router and now I've got remote access into my LAN. My laptop has essentially become a Linux server. I just ssh in from the phone and write code. Can also easily launch and use codex.

I have been using docking stations since 2006, and apparently if I want to buy a serious desktop that doesn't look like car tuning in a box, I have to build it myself, given the available configurations on ready to go desktops.
I went some time without a desktop computer at home. The idea was, I have a perfectly good laptop, I'll just connect it with a dock when I want to use a desktop-like device.

Never ended up happening. I didn't get much done at home at all and what I did get done was mostly on the couch; not a great environment for serious work. (I have an office where I do for-profit work so it's not an issue, but still, I like my side projects too)

Eventually I had enough decommissioned computer parts that I could assemble them and re-commission them into a working desktop. So I now have a desktop again. And I actually end up sitting at the desk working on stuff on the desktop in a way I rarely did with the laptop.

every time i try to 'streamline the computer setup' i get reminded that losing 'the desk' was the worst part / saddest moment
Is there an option to "use the remote computer just as it was local" even between two Macs? Sure, terminal/ssh is enough for most use-cases, but this is something I expected from Apple as builtin.

I can tailscale into my home network, but what next? VNC/TeamViewer/Remote Desktop setups are all mediocre imho.

Astropad Workbench
I once tinkered with the game streaming set up available via Steam, so I could do it in a way where it streamed the normal complete desktop setup. I used it to operate my desktop from the laptop.

It worked relatively well, but wasn't perfect. It was as close to what you're describing as I've seen, though. This was a few years ago; something like this might work even better now.

Unfortunately there is nothing that works as nicely as Windows Remote Desktop. Probably the best option is various commercial offerings as Apple Remote Desktop is rather buggy and unstable.
Apple screen sharing works for me just fine. Matter of fact I have been driving a mac mini with it from my laptop most of the time now and just ordered a mac studio to replace the mac mini. I think powerful headless desktops are the future.
I'm writing this right now connected to my Mac Mini from my "thin client" M1 MacBook Air. Screen sharing is pretty seamless nowadays, particularly if you're on a local LAN (although it works pretty well over Tailscale as well if the WiFi is strong enough).

If you haven't tried Screen Sharing since they deprecated VNC and switched to their proprietary H.264-based protocol, its worth trying. Even YouTube videos play just fine with no noticeable lag.

Thanks! I didn't even know this existed!! :o
Well, there used to be nxhosting back in the NeXT days, which I _really_ miss.

Unfortunately, it relied on Display PostScript and Quartz née Display PDF isn't architected to allow that sort of remote display/access on a per application level.

I find Jump Desktop to be excellent. One time payment. Can connect from iPhone / iPad, too. A noticeable step up from Screen Sharing (which can’t be used from mobile device anyway). Can’t recall it ever failing to connect or flaking out as so many RDP setups seem to.
Apple Screen Sharing is probably the best when it's purely within Apple products. But if the mix is more heterogeneous, I've found Sunshine (server) and Moonlight (client) is great.

This open source software was built for lag-free gaming over a network, but I've found they're great for desktop usage too. I frequently use it to remote into my Windows tower. Feels about as responsive as if I was sitting at the desk using the tower directly.

I switched back to a desktop tower a few years ago as well. Notably, I don't use external displays with the work-issued laptop, but rather set up the laptop on a stand alongside the dual 1440p displays and have synergy for mouse and keyboard sharing.
Man I did exactly this with the PC I built last April, it was the best decision ever. I love having a dedicated space with a huge tower running everything. Two big monitors mounted to the back of the desk on arms, huge mechanical keyboard in the middle with a big ergonomic mouse pad and hand shoe mouse.

It’s a whole “thing” when I go to the computer now. And frankly, it’s made it way more fun to use. It’s kind of like enjoying the process of listening to vinyl rather than pulling up a song on Spotify

As long as we keep going back-and-forth and buying stuff, trying to figure it out, the economy will be in good shape.
You don’t need to spend all that money to get your mojo back, make the changes that you would make I.e. leave the MacBook on your desk and see if life improves.
That’s nearly what I do but on a smaller scale. My iPad Pro serves as my laptop 90% of the time, and the 10% of the time I need to actually code and test in a chromium browser I remote into a mini.
Having replaced my MacBook with a Mac Mini, I would reconsider. The MacBook is just such a _complete_ package. Great speakers, great keyboard, fantastic screen, the fingerprint sensor thingy.. Takes a lot of gear to match that
Not an option if you're trying to drive 3 decent screens.
Macbook (base) yes, but my MBP (M3 Max) is driving 4 screens for me without issue.
My M1 Pro drives 4 screens no problem. The trick is that you can only attach two through one Thunderbolt port, so you have to attach the 3rd one via the other port instead of your dock. (The 4th screen is the laptop's built in screen)
Keyword is pro. Non-pro chips drive fewer screens unless you use DisplayLink. Which I did at some point, it's a hack but ok for lighter usage.
Yes but if your debate is desktop or laptop, you're probably not looking at anything but a pro.
I've been away from Macs (2015 MBP and an M1 Pro) for a while now, but they used to have a limitation of one screen per port. Apparently driver related as according to my faulty memory a Linux install didn't have the same issue.

Are you using Apple displays or did they fix it?

I do remember the M1 chips being uniquely capable in this aspect compared to later chips but when I say decent displays I meant 3x 120hz 4k screens with HDR on at least one.

Last year when I was browsing the only recentish Apple silicon capable of driving that was the M3 Ultra.

Honestly the screens become more of a liability than an asset to me past 2 (including the laptop screen). I've tried. Even if I'm knee deep in work and have like 4 servers I'm monitoring, my eyes are only going to look at one screen at a time.
I thought the question was low-end MacBook + mini vs high-end MacBook. The low-end one has all those nice things. I wouldn't sacrifice that.
Except for LLM inference speed, macs suck as a server and are pretty expensive, why wouldn't you get a Linux box?
Because it's not exactly a server, more like a home PC that I also remote into. Usually SSHing but sometimes VNC which is annoying in Linux. And the Mac mini is pretty fast. Yes a Linux box would win in terms of multicore CPU or GPU. My work is running those on cloud, not in my house.

I've never had a Mac Studio use case. Guess it's like mine but yes heavy processing, possibly local ML, where it can be cost-effective depending heavily on what you're doing.

I also have a new RPi. It's still too slow as a PC or server, and the random issues with Linux software on ARM aren't worth. Like oh, there's no arm64 bin for this thing so you gotta build it, and the libs are harder to find, and oh turns out the code has a race condition that only gets triggered on arm64...

Buying a Mac specifically for serving is misguided in nearly all use cases, but as an all-in-one solution they make sense. Personally I feel like the number of boxes I need to manage has an inverse correlation to my happiness.

Macs are totally “fine” for light server duty… as is just about any computer of the last decade+. The CPUs are beasts, the disks are screaming fast.

The operating system itself may not be ideal at serving but you can just run Docker/Orbstack if you need to do something especially Linux-y.

I’d put the question back on you — what are scenarios where an Apple Silicon Mac wouldn’t cut it as a light server for one person or a handful of people? About the only scenario that comes to mind is scenarios where you expect to utilize it so heavily that the fans are running for many hours a day. At some point those are either gonna wear out or just ingest so much dust that the machine runs hotter and needs a deep clean. But even that is largely mitigated by just pointing an external fan at it.

My Mac mini m1 is a fantastic home server.

Years ago I realized running a 100W PC all the time was REALLY expensive, so I switched to an old linux laptop at 30W (about $4/month). That helped, but I moved to the m1 mac specifically for the power efficiency. It does all the things my linux server did, and it does them at 6W (75¢/month).

I've found no competitor with similar performance that can run in such a low power footprint. M4 mini's are way better at power/performance, and I imagine their price is about to come way down since the M6 mini just got announced.

Software-wise, it's different, but mostly equivalent. Homebrew or macports has a similar software inventory to debian. And Apple's container framework is a welcome improvement over colima for running most container workloads.

That too, it's either Mac mini or Rpi if you care about power, and the Rpi is very limited.
My home server is an ASUS NUC 14 Essential which idles at 5W with a bunch of services running. Its peak performance is probably less than the Mac Mini, but the hardware is cheap (at least it was before RAM/storage got expensive) and it’s a great platform to run Linux on.
NUC is good too. Having x86 is surprisingly important in Linux.
It takes almost 8 years for $3.25/mo of electricity savings to break even with a $300 difference in up front cost. Assuming the time value of money is zero, that is.
Yeah, the reality is you never break even. It's only a thing if the two options are similar pricing.
I'm thinking the same, except I'd never go Apple again. I am planning to get a regular, modular mATX desktop (+Linux) for vast majority of my computing needs and strongly prioritize focused work at my desk.

It's (relatively) cheap, powerful and as problem-free as it gets.

I'd never go x86 again after owning an m1, and I have a, I guess now, $5k+ 7900x + 4090 sitting next to my 7 year old mbp that I would have had to replace with 2-3 x86 laptops by now
I can't wait for an actual competitor to the M chips from apple. It's frustrating.
What CPU / GPU combination would you recommend? Can use unified host+device memory?
For a desktop you may find yourself better on a Linux or Windows machine price/performance wise.

I personally own an M3 ultra, an M1 max as laptops, but my desktop is a Ryzen desktop I built in 2022 and it was a third in price of the ultra for more power.

Recently took a Minisforum 7840hs PC out of rotation as a media PC and made it a full time coding workstation with Proxmox. I do a VM per project due to the nature of agentic editors.

I was using a VM setup on my MBP but it felt like a huge waste, having to leave a laptop on 24/7 when all it did was run Claude Code inside VMs.

I likely will stick with a Macbook Air 15" for next purchase, and beef up my "Claude Server" down the road.

Yep, this is my exact thought. The pendulum has swung back toward a desktop making more sense for me than a laptop. It all depends on whether there is anything useful to do with an amount of computation that can't be fit into a laptop package. For a long time there wasn't, now there is.
This is my setup (except I have an Air because the Neo didn't exist yet). And it's great. Tailscale makes it trivial.
So, do you use VNC (or ssh) to access the Mac Studio at home? Or do you only access other stuff that's living in your home network?
ssh in terminal and/or through VS Code. When needed, I also use Apple's Screen Sharing app.
>I do wonder if my next computer should be a Mac Studio instead of a MBP that lives its life docked.

Having owned 3 MacBook Pros since 2008, the decision to make my next computer be a Mac Studio came down to (1) MacBook thermal throttling that slows down CPUs when it starts to overheat and (2) easier upgrade of Mac Studio SSD with after-market storage module whereas the MacBook requires more complicated disassembly and hot air gun to dislodge the surface mounted SSDs.

I have a brand new M5 Pro MacBook Pro I don't like it when the fans turn on. The Mac Studio will be faster and quieter for the same workloads.

You don't want a computer that does thermal throttling but also do not like it when the fans turn on?
I used to game on my 2014 MBP. That was a bad idea - took about 5 years, but eventually my batteries became spicy pillows.

Turns out that even with the fans at full blast, the batteries didn’t like being so hot over prolonged periods.

I’m happy with my M4 mini now, with a separate windows pc for gaming.

I have been thinking a lot about buying a very fast desktop and a cheap laptop that uses remote desktop to connect to the fast computer.

But in the end I ended op buying a Lenovo Legion and put Linux on it.

Laptops are so fast these days that I didn't want to be bothered with setting up connectivity to a remote desktop.

But if your laptop never leaves your desk I think a desktop computer is a great option. Relatively cheaper and easier to maintain and upgrade.

At my current job, one of my biggest blunders was thinking "Let me just order the same hardware as most of my teammates to avoid unnecessary complications".

Most, if not all, of our current work happens on remote cloud vms. Now I'm stuck with carrying a 3KG monstrosity to work.Every day.

Absolutely no positives compared to my ThinkPad that weighed less than half in my previous job.

Remote development is "good enough" these days. With VS Code, Development containers etc... Having a light weight, portable laptop is so much nicer than a laptop that you can't even rest on your lap for long duration.

The only thing you need to be mindful of using light weight laptops is not having enough RAM to fit all your browser tabs..

Yeah I stepped down to a smaller MBP. the only reason I didn't go for an air is my home setup with dual monitors involves using the HDMI port. There probably is a dock or something out there though.
I’m doing essentially this, and got a MacBook Neo.

Fir kinda the first time in my life I don’t really have development tools on my personal laptop. I ghostty and openvpn client installed.

I have a large remote linux workstation (2x 8c/16t xeon cpus, 256gb ram, 2x8tb spinning rust disk) and i have my tools over there (along with some VMs).

It works surprisingly well.

Also, the macbook neo is a surprisingly capable little machine.

(comment deleted)
Apple prices have always been insane, there is a reason why during the days it almost went bankrupt, in Europe it could not rival with PC, Amiga, Atari, Acorn.
Another option is a 'maxed out' mini e.g M5 Pro/64GB. Mini is super easy to travel with if you know there'll be a 'dock' at the far end e.g office<>home or whatever.
A part of my brain can’t process acquiring a Mac Mini M5 Pro knowing that they sell M6 despite being aware that it is just a weaker processor.

It will get worse with the MacBook ultra rocking a M5 too.

Apple’s marketing department is gonna kill me.

Think about it maybe like:

I want a 2026 pickup truck. The dealer also sells a 2027 hatchback.

I am not shopping for a hatchback.

I had the same view recently. For the past year I’ve been using my iPad to remote into both my MacBook Pro M1 Max and my racked Linux workstation at home using Jump and moonlight/sunshine, respectively. Never looked back
My goto. Even ignoring speed, it's nice to have a remote machine that keeps doing whatever it's doing while your laptop is closed. And the ARM Macs idle at such low power that it's not wasteful like my MacPro4,1 was haha. My UPS's ammeter doesn't even display the Mac mini's draw.

They gave me a nice MBP for my new job. I tried doing heavy work on it locally, it was fine for that, and yet it still ended up being a light terminal into an EC2 instance. And yes 50% of that is just having Claude not get interrupted, but there's tons of other stuff I want persistent.

Yes, except that the only way to get a fancy screen is also to put the fancy cpu in it. If there would be an Air with the screen of the pro I would buy it. Maybe even a Neo with that screen.
Oh yeah. Throw it behind tailscale. Herdr multiplexer in terminal is cool, it’ll remember your session if you open it in ssh from the neo
I would not use Neo, Air with 24 GB ram should be your minimum because you will want to do some work locally even when you're running a VPN/SSH thin client setup.

I'm currently SSH-ing into my workstation from my M4 Air with 24 gb ram and it's ideal for this flow. Slack/editors/clients/browsers/etc. easily gobble up over 16 GB. I have no more dev tools/compilers/source on my client machines, everything is dockered up on remote workstation in isolated VMs (too much supplychaining)

Can that run qwen? Also can you explain what you meant by the downside ?
If I was going to run LLMs locally I would definitely go for a pro.

I meant the only downside of not using a pro for my use-case is that I don't have a HDMI port on device with high refresh rate. I have a decent dock that can give me 4k/60hz HDMI but not 120hz refresh for macs (it works on windows). It's a minor thing but having 120hz is nice when I'm docked at home.

It heavily depends on your workflow and how many (electron) programs you need to run simultaneously. I found that if I’m mindful of closing things I’m not actually using, and let macOS do its swap magic, the Neo will just work.

I’m also the kind of person who close tabs and I like to work on one task at a time.

I'm sorry but that's a bit ridiculous. I'm still using an M1 Pro with 16GB of RAM and I have quite heavy usage - docker, other heavy native apps, a bunch of VS.Codes and coding agents, things running in the terminal, background processes. It's just a big exaggeration that general regular productivity apps "easily gobble up over 16GB"

(Edit: saying this as someone who just paid for a new 128GB M5 Pro Studio -- but that's for LLM's, not regular work)

I have an identical gen mac mini I bought for testing out the apple silicon for my workloads a year and a half ago, except it's 16GB min spec version - I got ram throttled on that thing constantly as soon as I start using it for development. If it was just productivity I could get away with it - but as soon as I start something like JetBrains Rider on it or VSCode with C# language memory pressure is in the yellow and if I start a next server for frontend the machine just grinds to a halt. If it was more readily available on retailers such as Amazon I would even recommend 32gb ram air for local workflows - if you're doing development these days you're just going to need it - especially with LLM workflows and parallelizing work. The price difference between 16 and 24 is not even close to not justify picking it for a work machine.
Cheap Linux laptop with Tailscale back into the Studio when traveling?
No iTerm2, weird shortcuts, random issues in video calls?, also not really cheap to get a laptop that reliably works with Linux and has decent battery life
Im writing this from a MacBook Air.

Everything is a container or VM now, and none of it runs locally for me.

I have a desktop (older intel) with giant monitors and a keyboard for when I sit at the desk. I have the laptop for when I travel, go out or just want to work from the couch.

When I do my next upgrade to "better hardware" I'm not migrating a machine, rather I'm migrating the containers. My workflow is such that if I loose one of the boxes I sit at to a cup of coffee I really wont care other than the financial loss of a new laptop or keyboard.

The biggest win in all this was dumping the off the shelf firewall/router and moving to Opnsense. Wireguard vpn lets me route all my traffic through home for all my devices (and what is now a growing home lab).

There are scopes of work that this setup would not work for. I would not want to be a video editor with this set up, it's not ideal if you want to play AAA games. But for what I do, it is pretty ideal.

I used to have a high-powered laptop, but just use a combination of ssh and NFS to my machine at home from a low powered laptop for travel these days. It works well.

Makes managing both backups and handling failure scenarios involving loss or unauthorised access to the laptop less of a hassle as well.

Why not a non-mac cheaper desktop/workstation and a Mac laptop for use?

E.g. what most major tech companies have - a laptop that is for VSCode via ssh/web browsing, and a beefy Linux dev box you ssh into for everything else?

It's way cheaper, and what I use at home too - a lot cheaper than a Mac studio for everything, especially with RAM and storage.

A lot more upgradable too. Those empty DRAM slots can get populated one day!
The unified memory is the big appeal of a beefy Mac Studio
if you do want to do local LLM stuff, mac studio is significantly better than most linux dev boxes.
As of a year ago, the strix halo AMD (system on chip, unified ram) was within single digit percentages of the mac, but it’s a linux box, so much better for most dev and as an llm server.

I know AMD has announced the replacement, and it will be faster, but I’m not sure when it’ll ship.

Also, you can cluster strix halo if you need more than 128GB of ram.

Better than a DGX Spark running Linux? Absolutely not (I own both).

Real-world matmul is about 4x higher on a DGX Spark than a M3 Ultra.

For specifically unified memory reasons, Apple machines are much, much better at running local LLMs.
I’ve debated Mac Studio or MBP at each generation. For equal spec machines the price premium to get a MBP over the Studio is always smaller than I expected, so I pay the extra amount and get the MBP.

This changes when you get into connotations that aren’t available in the laptop form factor, but with RAM prices the way they are those configurations are more than I want to spend on a local machine right now.

For running local LLMs the high memory Mac options always look appealing, but the processing speed (prefill) is so much slower than GPUs that it hurts. For situations where you have no rush and can let something work in the background for 24 hours it can some times be ignored, but the speeds I get from a real GPU setup are so much faster that I never use the larger models on a high memory Mac any more.

I got an M1 Max studio w/ 64gb when they first came out and I love it. Price was obviously a lot more reasonable at the time. I’m fortunate to have dedicated office space, and there is definitely a nice psychological effect to having a desktop in a specific spot and “going to the office”. I’ve since purchased an m3 MacBook Air as my thin client and for lightweight travel use, and when my wife needs to take a call in the office or whatever I pop out to the living room and either hop on Remote Desktop or SSH into my Studio. Highly recommend this setup. If you’re primarily doing thin client on a local network, a neo should be fine for that (and to use as a media player/communicator around the house). That said, I have been eyeballing Omarchy…
Happy with M4 Mini, although I recommend using a well ventilated M.2 enclosure instead of the integrated docks as I just had a 2TB Kingston die which I'm pretty sure was caused by "zero thermal forks given (chopsticks only)" UGreen hub design. Also can use as a laptop and replace screen/keyboard/whatever when failed/unsuitable. Higher resulting device longevity, especially with Asahi Linux making such great progress. https://github.com/vk2diy/hackbook-m4-mini
I got an OWC Express 1M2 precisely because of thermal fears. The heatsink is huge, but no fans. Only real issue is needing to be careful to not nudge the cables.

I also suspect that trying to stream larger LLM models from disk hammered the USB 4 connection too much - lead to system hangs, so now stream from the internal SSD instead.

Cool product, thanks. Sent back for Kingston RMA, expecting a new drive soon.
I’ve wanted the idea of a home server or even home “mainframe” and some terminals for me and family forever but the idea never takes off….
Same. For me, it’s one of those “solutions in search of a problem.”
It just seems somewhat more efficient to take all the computing storage, memory processes, processors, or at least all the money that go into it… the in one spot.

If you’re running a big task or small, everything scales to the appropriate size, regardless of the hardware sitting on your lap or under your desk.

At least that’s how I think of it.

I've been using the free version of TailScale to use macOS's built-in screen sharing app to use my home Mac mini from my work MacBook Pro. Works great! That combined with ChatGPT/Codex's "remote" feature is an amazing combination. I think you'd be totally fine with a Mac Studio + Neo.
I have been considering this.

I bought a Macbook Air in the interm waiting to see what the Macbook Ultra looks like, but honestly—it's the best form factor ever. I love it! Even though it's only like a pound heavier, the Pro feels like a monster.

Thinking about getting a Mini or Studio with more horsepower to stay at home.

Any other apps or suggestions? I'm still not exactly sure what working this way looks like.

I've been running a desktop Mac in addition to MacBook and iPad for the past decade (currently a first gen Studio still holding quite strong). It's primarily a focus thing, secondarily it's a way to keep work happening locally while I'm elsewhere, and never have the frustrations of docking and screen geometries all re-arranging, etc. The iPad is at the opposite extreme, it's for recreation, burn out management. The MacBook exists for when I actually need to do "real work" outside of the house.
YMMV, but when I got my iPad I realized that my "mobile" needs were only contempt consumption and my MBP was used as a desktop computer connected to a 32'' monitor. So every time I renewed my laptop the expensive screen went with it, and I had to pay for another one that I would be using as a second screen!

Thus, my current workhorse is a Studio.

I almost bought a studio a couple years ago, but I was starting a PhD and so I figured I had to prioritize MBP.

These prices make that decision so bittersweet. I feel so shut out from being able to use AI as economic productivity without being able to take on debt for a local-inf capable mac studio

Last year I was able to justify a 128GB/1TB M4 Max Studio for 4600€. This cycle's M5 Max with 128GB/1TB goes for 6219€. The M5 has to be obviously a better machine, but it's now above the price point I would pull the trigger for. Guess I got lucky in the futureproofing lottery this time, anyway.
> I realized that my "mobile" needs were only contempt consumption

Is that the new phrase for social media? I hope it is...

> not that Apple prices weren’t insane before the ram/ssd shortages

Funny that after the price started to increase last year (I think October/November), there was a window of a few months where Apple prices stayed the same as they were before, thus making apple prices actually good when compared to the rest of the market.

It was a unique opportunity to have acquired a 512G M3 ultra for $10k.

not as crazy a config, but I got a 128GB M5max the day before the price increase for $2K less. still proud of the call to buy it.
Also proud of having spent $2.5k on a used 128G M1 Ultra back in September 2024.

Despite being outdated in terms of compute, it stills let me run very good recent models locally, with Deepseek V4 Flash 0731 being the greatest one right now, and hopefully Qwen 3.8 Flash will also fit well when it is released tomorrow!

The M1 ultra definitely leaves to be desired in terms of its token speeds, but I think 20 tps generation and ~200 tps prompt processing (which is what I get with DSv4 flash), is already enough to do a lot of serious work when you combine with the decent prompt caching provided by llama.cpp.

You're killing me here lol
I have a Mac Studio M2 Max as my main driver. It's fantastic. I will get the best M5 Studio I can afford when the higher RAM options come out because I am very involved with local inference. I find that with Tailscale I can either use screen sharing if I've got a good enough internet connection or just SSH and Mosh if it's a little bit laggy. That has been good enough for me. A Mac Studio and a Neo as a companion is a really good combination. My laptop is an Apple M4 Air but it basically does nothing except for running ScreenShare and a terminal.
> I do wonder if my next computer should be a Mac Studio

Would you get away with a Mini?

Slightly off-topic: I decided to go with a Mac Mini to start playing with AI. My daily driver is a Linux laptop (with a negligible GPU). I also have a few Raspberry Pi machines for backup, and a low-end AWS server.

I was always cobbling together ad hoc network access, to get from one machine to another. Access to my Mac Mini while traveling was a pain. Bringing it with me is ridiculous, and network access was a PITA.

Tailscale is wonderful magic. Free (for my usage), so easy to set up, and now no matter where my various computers are, they are all accessible trivially via a single ssh connection.

So get a beefed up desktop Mac, set up Tailscale, and then use any random laptop, anywhere, to use it headless.

Yeah, I have an MBP for Reasons, but it sits docked 90% of the time. I'd rather spend the engineering budget on better thermals and more ports; I always have it plugged into a fixed monitor and other peripherals.
I’m in the same camp as you. I’ve just decided that the M5 Ultra is gonna be my next machine. Plus an Air for the few times when I need a computer on the go. I’m invested in Tailscale already so an ubiquitous Studio Ultra seems like a sweet idea
I don’t even know what a Mac Ultra looks like and I want one.

It will be new Mac with more GPU cores for ai, right?

It would sit in my basement server room providing local inference.

I have exactly this setup and use a Macbook Air with Tailscale to just connect via High Performance Screen Sharing both at home and on the go (when not sitting at my desk, I move around a lot around the house after 3pm or so while I keep my workflow identical)

HPSS works well via 5G, and Tailscale is incredible for the setup as much as the hardware.

Edit: I ended up getting a 15" M5 Air but honestly trying it on a 13" M4 Air was actually superior because when all you care is mobility (since the Studio does the heavy work) its nice just going around with almost no weight / bulk.

M4 Max Studio + Macbook Air. I've been running this for several years. Mac is a really poor server environment and none of the apple ecosystem makes up for it. So, I still prefer my linux server, but the M4 Max is much more suited for inference tasks so here it is. I wouldn't say it's much better than a MBP except the sustained throughput is much higher and the fan noise is barely noticeable. I'd get a linux equivalent if there was one.
I used laptops for years.

However, after retiring, I realized I never undocked my MBP.

So I got an M4Pro Mini, and I've been thrilled. If I ever get to where I travel a lot, again, I'll get a laptop, but I don't see a need, right now.

I can't really imagine buying a new computer in this market. But I also wonder how long it will be before prices drop again, if ever.
If it's your remote-in laptop, I'd suggest getting a used MacBook Pro or Air with an M series processor in your preferred 13/15" size and memory/storage configuration. You can get a used MacBook Air M3 with 16GB of RAM for about the same price as a MacBook Neo. It'll have similar single core, GPU, and NPU performance as the Neo and about 38% better multi-core performance. Plus you'll have 16GB of RAM for future-proofing and apps. And you can spend a bit less if you want or a bit more depending on what you need. You could get a used MacBook Air M1 8GB for under $400.
I wonder if apple could or would offer a subscription for a OSX VM that you could remote into? That plus a neo would be an excellent combo for me
Yes, and..

Just Tailscale into the Studio from iPad Pro 13" with magic keyboard and Kit Knox's rootshell:

https://github.com/kitknox/rootshell

Note that the iPad Pro can also drive a 4K second screen if you like, and most anything else a Macbook with a single port could drive.

Why not Macbook Air? Because the iPad Pro is also a tablet, touch screen, and 5G...

thanks for this link! I was constantly struggling with remote-ipad setup - tried so many things (including paid options), but everything was _iffy_ at best (on-call4live, lol)

And I'm also gonna ride my M1Max Studio until it dies (hopefully it will outlive me xD)

Yeah, I was about to say "or an iPad Pro", but the Neo is shockingly cheaper than an iPad Air.

There's definitely an option for everyone if you want to go down the "fixed supercomputer, remote terminal" route.

Why use mac at all? If you want a docked machine you can build a PC with the exact specs you want. Throw a Linux distro on it and you will have a similar experience
Because it's NOT the same experience. iCloud access, shared messages, Cut and paste between other Mac devices..I say this with a Ryzen 9 + rtx4090 super right next to the macs. (A studio and an M1 MBP with a new battery).

While they're theoretically at parity, in actual use, they're not.

This was my perspective for a long time. The issue is the Macbook Pro screen is awesome, and the display story for desktop computers is pretty grim.
10 grand for 256GB memory. Likely double that for 512GB, but won't be available or finalized until October. Thunderbolt 5 is highest bandwidth external IO available at 120Gb/s. 1.2TB/s claimed max internal memory bandwidth.

Not exactly "future proof" for >1T parameter models but good for targeting specific lower-parameter models, or if you can rely on pipeline parallelism and run a cluster.

10 grand for 256GB new Ultra sounds too cheap in today’s crazy DRAM market, it feels too good to be true.
There is no "future proof" for >1T param models, there is no present or future where you can run a model that size on consumer hardware.
6 Mac Studios for 100k USD. Consider it a rule of thumb now - 100k to run 1T params, scales linearly.
Out of curiosity, what do you do to distribute the weights and inference across the multiple machines?
Thunderbolt 5, and software
at a 100k$ you're not buying apple products to be limited by the 120gb/s thunderbolt port.
If you have 100k you can just buy a few b200s.
One does not simply "buy a few b200". They come in 8 packs minimally, at >5x that budget.
I don't know that "consumer hardware" is a useful distinction anymore, it's just "what's your budget and what's your speed requirement".
Consumer hardware means: - 120v input plug - not rack-mounted - has a video out port
Ah, a MacBook Professional is Consumer /s
(comment deleted)
Apple's use of "Professional" is just a fancy alias for high end or premium, it in no way indicates anything about being "for professionals" (most extreme example: "professional" iPhone models)

Hence why they had to make up the "Studio" brand for the workstation market, because they'd already fully removed any meaning from "Professional"

In computing "Pro" is not the opposite of "Consumer".

Putting aside the fact it is a marketing label, "Pro" usually means "designed for work" while "Consumer" (in this context) means "doesn't need a special environment".

In computing the distinction is primarily noise, power and cooling requirements.

If a computer is designed to use home power and is quiet enough to use without annoying people and doesn't require specialist cooling then it is a consumer device, even if it is used for work.

To me, "consumer item" is when i don't have to contact sale people for price.
That's in the same ballpark as two 128GB AI machines like the Asus GX10 or DGX Spark or Strix Halo. And, it seems very likely to perform better than either of those for inference. And, 256GB brings some pretty good models into play.

But, that doesn't make it a good deal. It just means the Apple tax doesn't apply when stacked up against AI machines and with memory prices being so out of whack. I'm still planning to wait until the RAMpocalypse ends before I buy any more hardware.

Is there anything in the works or planned that suggests the RAMpocalypse will end anytime soon? e.g. new fabs being built, permitted, planned, etc...
There are signs that the money faucet is being turned down. Various investments that were announced have been quietly canceled or reduced in scale. I don't think it'll happen soon, but it seems unlikely to be more than a couple years. If I were a betting man, 12-18 months seems right. The AI companies that can make enough money will survive, the ones running on investor cash and debt, won't.

Efficiency is improving, both in hardware and in software and in intelligence density (smaller models can effectively do more of the AI work that needs doing), so I think the pure data center plays will falter. If there isn't some other business attached, they're never going to recoup their investment. Anthropic and OpenAI are buying all the compute they can find right now, but efficiency gains, especially those coming out of Chinese labs where they must be more efficient to compete, will make it less and less of a problem.

I mean, think about the hardware we use for AI. It's basically an accident. GPUs were not designed for AI (though they are becoming more focused on AI). The specialized AI hardware industry is just ramping up.

So, we're still early in the curve for how efficient both the hardware and software can be at performing these tasks, and given the effectiveness of recent very small models (e.g. DeepSeek V4 Flash 0731 and Qwen 3.8 27B), I just don't see a long future for giant data centers built around billions of dollars worth of last years graphics cards. As with the crypto mining operations, at some point, it becomes more expensive to run the hardware than it makes in revenue. And, as with the crypto mining operations, when the money dries up, the hardware hits eBay and prices drop.

There was an article last week that a Chinese RAM manufacturer was planning to add new fabs to be able to ramp up. More or less simultaneously there was also an article on Apple considering switching to using china sourced RAM for China destined devices.
Define "soon"

RAM production is completely sold out for 2027[1] which means the prices are locked in until after then.

It takes about 2 years from the time ground if broken for a new fab to be built and producing RAM.

There were some new fabs announced between February and April this year by both the Korean and Chinese manufactures, so that new capacity might start having an impact in 2028 in the most optimistic scenario.

Samsung says supply will remain tight in 2028[2], and Micron says "tight beyond 2027"

The best hope is that new (Chinese) players overbuild fab capacity and supply outstrips demand. That isn't likely, but perhaps in the late 2028-2029 timeframe could happen.

[1] https://www.techpowerup.com/351344/memory-makers-seal-2027-d...

[2] https://www.tweaktown.com/news/112966/memory-shortages-will-...

[3] https://s25.q4cdn.com/621799436/files/doc_events/2026/06/Q3-...

No one making statements has an incentive to tell the truth though (in fact, they're all incentivised to keep proclaiming the RAMpocalypse will never end).

Producers (Samsung, SK Hynix etc.) will not say "prices expected to drop" or "demand expected to drop" even if it was true because then consumers would start delaying purchased.

The big buyers (OpenAI, hyperscalers etc.) have no incentive to say "supply expected to start opening up" because that would imply their growth trajectory is flattening; also a huge part of their moat now is just deployed RAM.

There are some news about AI/LLM progress rate flattening out: People uses the cheaper model more than better model. Some AI startup in Chinese lose half of value.
The memory bandwidth on the Spark and Strix Halo is a fraction of the M5 Ultra. "Perform better" is probably a huge understatement.
Yes, if I was going to throw away ten grand on a computer that can run models that are much worse than what I can rent from a variety of providers for a few bucks a month, I'd buy the Apple.
It will always be wasteful to have a GPU sleeping next to you 99% of the time. Something like OpenRouter is the solution imho, for me at least. I realize that that some people care about privacy more though, and I respect that.
I wish they were offering 1TB of Unified Memory for the M5 Ultra. I already have an M5 Max MBP w/ 128GB of RAM for running local models, and while there's a /few/ models that I can run in 512GB that I can't run in 128GB that are interesting, where things really shift is at 1TB of memory which allows you run >1T parameter models w/ 4 bit quants reliably. 512GB is just on the edge of "enough", which is maybe the point of maximum frustration considering current memory prices.

Personally, I can't justify dropping the dosh for a 512GB M5 Ultra, but I would be able to justify it to myself if I could get 1TB of memory, because it'd guarantee the flexibility with local models I currently am missing. Seems a huge miss to not offer this... for a price.

1tb would likely be ~$20k - given the current >$10k price tag of 256gb. Would you still be considering it at that price?
chaining an option?
They say you can cluster up to four with a shared memory pool, and get three times the inference performance of a single machine.
It seems super reasonably priced to me. It's only twice as expensive as my first Mac which only had 128K of memory.
That came with a monitor and floppy drive.
and this one comes without a floppy drive.
Base models are okay-ish.

But +4000$ for an additional 128GB of ram is simply milking the customers, as they know they will have many of them.

Still absurd, but it's $4K for an additional 164GB RAM.
Pretty much every company milks their wealthier customers and it frees them up to be more competitive at the lower end. Also, if they charged a more reasonable amount, they would get backordered very quickly.
Depends on your perspective. People think nothing of spending 50-100K on a car that basically gets them to work. But the thing they use for day to day work then gets the evil eye when it costs more than 1K. It's slightly irrational. Not everybody needs a high end mac. But when you do, it sure is nice that you can get one.

I don't actually own a car and my startup is bootstrapped and our salaries are modest. But the one thing we spend on is laptops. I have M4 max pro with 48GB. That thing was on the expensive side (~4.5Kish). But it delivers a lot of value and I spend most hours I'm awake using it. I like fast builds. I like that I can try out open source AI models. And I like just having the option to run those.

We actually lease them and mine costs something like 105 euro/month. Including Apple Care. I don't need a Mac Studio but I could see some roles where that would not be a crazy expense. Even the tricked out version that basically only costs the same as a very modest car.

It looks like speculation that Apple would raise the base chip’s maximum RAM from 32GB to 48GB was wrong.

Apple also launched the base M6 today with a 32GB RAM limit, suggesting 512GB may remain the maximum for Ultra chips for some time. Since these Ultra chips combine 16 base chips:

32GB × 16 = 512GB

They probably literally don't have enough NAND to go around. 768GB of memory (48GB x 16) is enough for nearly 100 iPhone 17s; that's $800k of iPhones at MSRP, although likely much lower margins than these high-RAM boxes.
They also cancelled availability of the 512 M3 Ultra months ago in many regions, likely just redirecting memory
800 USD/iPhone x 100 iPhones = 80k USD
Oops, thank you, corrected.
they`re skipping m6 pro and ultra
AFAIK these "rules" are artificial and there's nothing stopping Apple from having twice as much RAM. They just don't want to.
(comment deleted)
Very much happy with my machine.Got a base m4 max studio in December. 2350eur what a deal
I was looking forward and hoping that the Mini and Studio would have 8K at 120Hz. Oh well, maybe the M7s will have that.
No looking to cheat, which is better: the MAX or ULTRA?
Here's me trying to justify this when I can run frontier models in the cloud for less than the monthly finance charge for this beast.
If you like to experiment with training / finetuning / etc on LLMs, these are actually incredibly ‘cheap’.

1.2TB/s memory bandwidth unlocks a lot with 256GB unified, and agentic AI is pretty good at optimising performance.

For comparison, to get 256GB with NVIDIA, you’re looking at a DIY workstation build (need pcie lanes), and like $70k?

The spark’s ~250gb/s bandwidth doesn’t really count here.

Neither is the right alternative to compare to. You aren’t going to hit 100% utilization (if you are, ignore me, this doesn’t some to you, and write a blogpost for me to read and share).

The comparison should be against renting in the cloud for the duration of your task for training and research or using pay-per-api-call providers for general inference instead of buying your own hardware (and paying the electricity and cooling bills on top), because let’s face it, the models you want to use are probably the same ones available on inference providers (but, yes, some are more trustworthy than others).

Speaking as someone that does ML/AI research, you are essentially paying a huge premium for being able to just run your Python script at any time without setting up a deployment script and harness to run the job remotely, while your hardware sits essentially idle the rest of the time.

The only way to make the math work is if you rent your hardware in the background for inference while you’re not using it in anger, but despite all the startups and promises that has never become as streamlined as mining bitcoins or shitcoins used to be and they don’t pay out as much as they say they would. Renting your hardware for training is another option but doing that is a lot more involved, options are fewer and farther in between, you won’t get as much utilization out of it, and doesn’t let you feasibly abort running tasks at a moment’s notice.

My card (RTX Pro 6000) is always doing something all the time from my queue; like some synthetic dataset generation up next. I still actively use runpods and openrouter for scaled stuff, I was spending a bit and then did the maths, and invested in it.

The maths to me was basically equivalent to prepaying for 242 days of runpod pricing for the same GPU; and I reckon I'd be able to get 6+ years of use out of this card with 96GB.

Plus there's the resell value -- it's actually appreciated by ~50% since I bought it.

Plus I do really enjoy that it's 100% local. I wouldn't feel comfortable giving my agents this much information if inference wasn't 100% local.

I wouldn't get another one, I wouldn't have as much value, but one is definitely paying off for me on the financial side.

Yeah not really lol.

Only reason to buy this if you want to own your compute.

Experimentation and inference are all going to be cheaper on the cloud

No, you need more compute for those use cases. That's why everyone trains on GPUs.
If I were to get one of these, realistically what is the most advanced AI model I could run locally?
GLM 5.3 in ~q4 quantiziation with 4bits
"512GB memory option for M5 Ultra coming late October"
anyone know if this is "pre-order" is coming late October, or "will be available to ship in late October, thus pre-order will be available earlier than that"
The article's headline contains an em dash. I wonder if it's AI-generated
Boy, oh boy Apple is the new shovel seller during AI gold rush.

Most people at Apple have already realized that their processors are already too powerful for regular users - heck, as a developer my M2 Pro with 32 GB RAM is more than enough for me.

Regular users don’t care about local AI either. So, they will probably extract as much money as possible during AI gold rush, but then we will most likely see Apple

a. Making their software worse (god forbid, forced updates)

b. Making their hardware impossible to repair (as they almost accomplished this already) and easier to break.

> Boy, oh boy Apple is the new shovel seller during AI gold rush.

They're eating nvidia's lunch.

The new mac minis and mac studios are going to be in shortage for at least first 6 months from 9.22
Am I the only one that now finds press releases like this similar to "AI Slop"

I know there's tons of marketing language, buzz words and attempts at convincing me of some agenda that isn't super clear without lots of effort in "validating" the slop. I guess its not bad "slop" though if a human put in effort in editing it (imo >50% human curating = not really bad ai slop)

Though I still would prefer I could just get the prompt. What human thoughts, direction and "prompt" went into writing this article? in the same way as we ask for the prompt for AI generated outputs, I would prefer it for human generated output too. For writing at the least. I could have saved time, got the purity of the argument, and got more clear information. I wonder if we can get a future where humans just express their intent and we can interpret and act on that without the noise of trying to impress or market to each other.

Sorry for being lazy, but is there a rough breakdown like "You get sonnet level for M5 and Opus for M5 pro, etc.", or is it still speculative. Or put simpler, do you get Opus level for the 256GB M5 Max?
For local LLMs with a Mac, rule of thumb is you always want an Ultra (due to memory bandwidth). Even an M1 Ultra is superior to an M6 Pro in this regard.

There are no configurations even close to running something comparable to frontier model variants, they're simply far too large, but something like full precision Qwen 35b or DeepSeek 70b at 50+ t/s is well within available configuration, and potential for plenty of room for large context sizes.

256GB is enough for DSv4 Flash, expect maybe ~30tg/s, and a lot better profile.

I'm using Flash heavily, and I would describe it as nearly as intelligent as Sonnet in agentic coding, but more usable and actually preferred model.

Takes less handholding, less likely to make unsolicited refactors or whatever, and the writing style is readable.

Even if you can fit the large models in RAM they end up being so slow I went back to the cloud models anyway
With the newest Qwen 3.8 27B model you can get opus on just about any new Mac.
It can fit into 16GB macbook ?
$5500 for 96GB of RAM? insanely expensive (the MacOs is really bad compared to Windows or Linux).