The stuff Hacker News actually recommends.

Desktops and mini PCs · Apple Mac mini

Apple Mac mini M4 Pro

24owners with an opinion
70%of verdicts positive
14/6positive/negative since 2024
0happy after 3+ years

Praised for performance and ease of use; criticized for reliability and value.

What owners wrote

Doesn’t like itOwned 1 year
It's pretty depressing as well as frustrating, watching something pretty solid get flushed in real time. This post puts words on it I hadn't really, but fading "confidence" is a really succinct foundation. There's no single massive thing just completely fubar'd, but the cumulative effect of hundreds to thousands of mysterious cuts combined with that now familiar sinking feeling of watching GUI bikeshedding accelerate with each new version, every more miserable, less useful, no more Feelings Of Delight at seeing some cool little attention to detail or nice new thing that clearly had some real thought and testing behind it. It all really eats away at one's day. I'm seeing this HN post and article literally right as I'm watching Activity Monitor and trying to figure out WTF is going on with mysterious ApplicationsStorageExtension (a plugin buried in the StorageManagement.framework) tasks spawning 4x on a Sequoia Mini M4 Pro and doing constant disk churn to no discernable purpose and introducing massive latency to certain file system access patterns. This started up out of nowhere in the last few weeks, almost exactly 1 year after getting this machine and with no major system upgrades. Why!? Who knows! Force quit them all and other stuff like MailStorageManagement goes nuts for a bit and then it settles down for an indeterminate period of time. Restart the computer and it goes away for awhile. No FS corruption issues found, seem to be no issues with spotlight. Sigh. After decades…
Neutral
I actually have a Mac Mini M4 Pro with 48G. I gave the k8s example because this is what I was doing with it. Was because I am back to using Linux as my workstation. My Mac Mini is now a headless server for llama.cpp. So, you are right that for these workloads , I would not be using the Mac Mini for k8s AND llama. Another thing going against using a Mac for Linux containers is that there are no solutions that I know that properly manages memory : memory is given to the Linux vm , but never fluctuates if the needs in the vm are less than the initial request. I know Orb Stack does that but is it proprietary. I think UTM does it , but not sure I would use UTM instead of Lima, Colima , multipass , etc to run containers.
Likes it
having a 64GB mac mini m4 pro the last few years with some increasingly capable usefulness has kept me interested in this stuff in a way that using a paid platform wouldn't have. Similar to running K8s in a homelab, something about interacting with the hardware makes it more engaging/interesting, for me at least. In general, I'm a big believer in doing more with fewer resources, within reason, and think having local setups really helps me be mindful with what's happening under the hood with these systems and managing context efficiently to get high quality results.
Neutral
"inclined to overthinking," holy cats you're not kidding! On a Mac mini M4 Pro 64GB I prompted it with "svg owl" and it thought for 17m12s, outputting 36.3KiB of thinking chatter. It did end up producing a 20.2KiB HTML+JS+SVG file with a very nice owl, including cursor-tracking animation, but it ran for more than a half hour! The MoE models are stupendously faster.
Likes it
Agreed. Here's a start for posterity. Using their prompt, "Write a 300-word explanation of how attention works in transformer models, aimed at a junior developer. Use one concrete analogy.": Note that my mini is the bigger Pro model, so it will be faster than the mini quoted in the article. Mac mini M4 Pro (cores: 10P/4E/20G) 64GB Tahoe 26.6 LM Studio 0.4.20+1 Qwen3.6-35B-A3B-MLX-4bit: 78.81 tok/s, TTFT 0.93s (but note Qwen 35B is really chattery and outputs 3,451 words of thinking for 65s first) Qwen3.6-27B-MLX-4bit: 14.65 tok/s, TTFT 0.76s (Qwen 27B output 3,027 words of thinking for 361s first, spinning up the fans) gemma-4-26B-A4B-it-QAT-MLX-4bit: 64.93 tok/s, TTFT 0.44s (Gemma 26B is much more on-task, thinking with 591 words for 15.62s first) gemma-4-31B-it-QAT-GGUF Q4_0: 12.25 tok/s, TTFT 1.65s (Gemma 31B thought with 404 words for 51.55s)
Likes it
I am not going to flag you, I am much OK with having good arguments. I just purchased a Mac Mini M4 Pro 64GB for $3k - 2nd hand of course. I am not a hater of Nvidia and I am planning on building a workstation based on RTX cards. You clearly do not seem to understand how convenient the MacMini actually IS - the form factor, how quiet it is, how durable it is, how well it integrates with other Macs, how well it works as a bridge to a personal agent like Hermes (integration with iMessage, Calendar, Reminders, iCloud, etc). I am pretty sure I know a thing or two about computing, I have been in the trenches for many, many years and I have had machines of all kinds, shapes and colors. It just so happens that Macs are very capable, very convenient machines that happen to work great in the era of LLMs, too. But you do you.
Likes it
It's amazing how far this emulator has come over the past few years. It runs so well on my Mac Mini M4 Pro system that it has effectively replaced my real 500 MHz Celeron rig for all but the most demanding activities.
Likes it
In my personal experience, 13” M1 MacBook Air, 15” M4 Macbook Air, M4 Pro Mac Mini, and MacBook Neo, the Neo is the fastest for single threaded and strictly CPU bound tasks. E.g. calculating 200x200, 1000 max iterations Mandelbrot fractals it does ~785 in ten seconds compared to ~760 on the M4s and somewhere in the 600s for the M1. Given its RAM size I’m not going to be spinning up VMs, but in terms of general purpose computing it’s more than adequate. And, out of the box, you get a word processor, spreadsheet, presentation, video editing, digital audio, web browser, and a bunch of other things. Xcode is free. This is easily a laptop you can buy and use for years in 90% of settings.
Doesn’t like it
I have an Mini M4 Pro with 64GB of 273GB/s memory bandwidth and it's borderline with 3.5-27B. I assume this one is the same. I don't know a ton, but I think it's the memory bandwidth that limits it. It's similar on a DGX Spark I have access to (almost the same memory bandwidth). It's been a while since I tried it, but I think I was getting around 12-15 tokens per second an that feels slow when you're used to the big commercial models. Whenever I actually want to do stuff with the open source models, I always find myself falling back to OpenRouter. I tried Intel/Qwen3.6-35B-A3B-int4-AutoRound on a DGX Spark a couple days ago and that felt usable speed wise. I don't know about quality, but that's like running a 3B parameter model. 27B is a lot slower. I'm not sure if I "get" the local AI stuff everyone is selling. I love the idea of it, but what's the point of 128GB of shared memory on a DGX Spark if I can only run a 20-30GB model before the slow speed makes it unusable?
Likes it
There’s a lot of commentary to the effect that the Mac hardware is good, but the software is somehow terrible. Speaking as someone who first used a Mac Plus, graduated to an SE/30, and is now on a Mac mini m4 pro… I can remember when Macintosh was widely held to be an acronym for Most Applications Crash, If Not, The Operating System Hangs. The software has always been terrible, until you try the other guys’ stuff. The hardware has often been good, and is in a purple period right now. Enjoy it while it lasts! (It won’t).
Neutral
I waited for such a device for years. I have 3 Macs at the moment. My MacBook Pro (still an M2 Max) from work for work. A Mac Mini M4 Pro from last year and an older M1 MacBook. I separate my work and private usage because of all the security software my employer forces on the machine. I didn’t want to buy yet another 4000Euro MacBook for my private stuff. So when the new Mini was announced I jumped on it. But my problem was that this meant that my personal computer was no longer portable. I was lucky because I still had the M1 in my house. My wife gave it to me because she stopped using it. If not for the M1 I would definitely have spend money for a MacBook Air or a used older laptop. This Neo would also fit my requirements. But I don’t need it at the moment. The second reason is of course kids. Yes I thought multiple times if I should get my kids some kind of computer. But the issue is that I don’t want a Chromebook or a windows device under my care. Used would be the way to go here but I love the fact that there is finally a good value entry level machine that also looks fun.
Likes it
This sounds more like problem of HP’s dock than a Mac. Just because they said it is officially compatible with Mac doesn’t mean it is. Also, compatible with which Mac- Intel or M series? I use three different docks on two Mac Mini (M4 Pro) and they all worked out of the box. I did my research before buying them by watching YouTube reviews.
Neutral
The 128 GB Strix Halo system was tempting me, but I think I'm going to hold out for the Medusa Point memory bandwidth gains to expand my cluster setup. I have a Mac Mini M4 Pro 64GB that does quite well with inference on the Qwen3 models, but is hell on networking with my home K3s cluster, which going deeper on is half the fun of this stuff for me.
Likes it
I also switched to cheapest Mac Mini M4 Pro this year (after 20+ years of using Intel CPUs). MacOS has its quirks, but it provides ZSH and it "just works" (unlike manjaro I used in parallel with Windows). I especially like the preview tool - it has useful pdf and photo editing options. The hardware is impressive - tiny, metal box, always silent, basic speaker built-in and it can be left always on with minimal power consumption. Drive size for basic models is limited (512gb) - I solved it by moving photos to NAS. I don't use it for gaming, except Hello Kitty Island Adventure. I would say it's a very competitive choice for a desktop PC in 2025 overall.
Doesn’t like it
Yes, that's why it was on the 3rd repair. Apple knew they had issues and replaced it for me before by replacing the entire main board. Twice. The last time I took it in, they would no longer replace for free and wanted $800 for the repair. That was half the cost of modern laptop, so I chose no. I was unaware of being able to disable the GPU like that. I still have it on a shelf, but honestly, I don't see trying to do the hack now but might have considered back then.
Likes it
The key benefit is significant lower power usage. Benchmarked llama3.2-1B on my machines; M1 Max (47t/s, ~1.8 watts), M4 Pro (62t/s, ~2.8 watts). The GPU is twice as fast (even faster on the Max), but draws much more power (~20 watts) vs the ANE. Also the ANE models are limited to 512 tokens of context, so unlikely yet to use these in production.
Doesn’t like it
macOS (OS X for those old enough to remember when it exuded quality) is no picnic either. Apple's software is arguably buggier than Microsoft/Windows these days. I haven't seen a real Blue Screen of Death in years, but my M4 Pro Mac Mini with 64GB RAM routinely hard-locks on a black screen when simply trying to reboot it, or starts beachballing with some process called 'audiomxd' eating my CPU.
Likes it
Been using Airs since 2011, 2012 model lasted me to 2018 upgrade (Retina finally) but now I snagged the m4 pro Mini and it’s so small I brought it to the birthing inn with me and (with the help of a universal remote) just used the room TVs as a monitor. It’s so small that I can just throw it in a bag with a mouse/keyboard/HDMI and even a PS5 controller which I do appreciate my wife tolerating Jedi Cal joining us in the postpartum wing over PS5 streaming from the console at home, quite doable and the Mini’s built in speaker is quite a charmer all things considered! I ran it from inside a drawer under the baby “kiosk” and it definitely outgunned the in-room speakers that were clearly gimped (similar to putting your phone into the right enclosure amplifies the sound). At home when not at my desk I’ve been using screen share to remote in from the 2018 Air, this is the first time since 2018 I bought a new computer and it’s oddly nice having it not be a laptop, don’t have to worry about the precious built-in screen or keyboard. Caveat may be if I wasn’t working remote perhaps it would be different but not sure, using the 2018 Air as a client for the M4 Pro has been pretty solid for my current purposes and it’s nice still having an Intel Mac for the edge case backwards compatibility development needs. Whoops didn’t mean to make a blog post in here…
Likes it
I just got a new M4Pro Mini (Apple). It replaces my M1Max 14-inch MBPro. Bit zippier (not screaming), but it does have native support for Apple "Intelligence." I was waiting for the M4Max/Ultra Studio, but, y'know, I realized that I have no need for that. This has been working fine, for a couple of months. I suspect that I won't be replacing it, for a few years. I probably will need to get a new iPhone, and maybe iPad, sometime in the next year or so (also for Apple Intelligence stuff), but I'm in no hurry.
Doesn’t like itHasn’t owned it
>it doesn't make any noise You can hear the fan at full load, especially on the M4 Pro. I really wish Apple went with a larger case and fan for that chip, which would allow quieter cooling. Also, many units are affected by idle (power supply) buzzing: https://discussions.apple.com/thread/255853533?sortBy=rank The Mac Mini is quieter than a typical PC, but it's not literally silent like, say, a smartphone.
Doesn’t like itHasn’t owned it
It's $800 to go up to 2TB from the 256GB model which is just criminally over priced. I can get double that for half the price with a Gen 4 NVMe drive. Weirdly the 8TB drive on the Pro is at least in line with the top of the line 8TB NVMe SSDs you can buy though there are cheaper options at about $600 vs the $1200 Apple is charging.