MudMan

A quick look at US Amazon spits out that the only 24Gb card in stock is a 3090 for 1500 USD. A look at the European storefront shows 2400EUR for a 4090. Looking at other assorted stores shows a bunch of out of stock notices.

It’s quite competitive, I’m afraid. Things are very stupid at this point and for obvious reasons seem poised to get even dumber.

Yeah, for sure. That I was aware of.

We were focusing on the Mini instead because… well, if the OP is fretting about going for a big GPU I’m assuming we’re talking user-level costs here. The Mini’s reputation comes from starting at 600 bucks for 16 gigs of fast shared RAM, which is competitive with consumer GPUs as a standalone system. I wanted to correct the record about the 24Gig starter speccing up to 64 because the 64 gig one is still in the 2K range, which is lower than the realistic market prices of 4090s and 5090s, so if my priority was running LLMs there would be some thinking to do about which option makes most sense in the 500-2K price range.

I am much less aware of larger options and their relative cost to performance because… well, I may not hate LLMs as much as is popular around the Internet, but I’m no roaming cryptobro, either, and I assume neither is anybody else in this conversation.

You didn’t, I did. The starting models cap at 24, but you can spec up the biggest one up to 64GB. I should have clicked through to the customization page before reporting what was available.

That is still cheaper than a 5090, so it’s not that clear cut. I think it depends on what you’re trying to set up and how much money you’re willing to burn. Sometimes literally, the Mac will also be more power efficient than a honker of an Nvidia 90 class card.

Honestly, all I have for recommendations is that I’d rather scale up than down. I mean, unless you also want to play kickass games at insane framerates with path tracing or something. Then go nuts with your big boy GPUs, who cares.

But for LLM stuff strictly I’d start by repurposing what I have around, hitting a speed limit and then scaling up to maybe something with a lot of shared RAM (including a Mac Mini if you’re into those) and keep rinsing and repeating. I don’t know that I personally am in the market for AI-specific muti-thousand APUs with a hundred plus gigs of RAM yet.

Thing is, you can trade off speed for quality. For coding support you can settle for Llama 3.2 or a smaller deepseek-r1 and still get most of what you need on a smaller GPU, then scale up to a bigger model that will run slower if you need something cleaner. I’ve had a small laptop with 16 GB of total memory and a 4060 mobile serving as a makeshift home server with a LLM and a few other things and… well, it’s not instant, but I can get the sort of thing you need out of it.

Sure, if I’m digging in and want something faster I can run something else in my bigger PC GPU, but a lot of the time I don’t have to.

Like I said below, though, I’m in the process of trying to move that to an Arc A770 with 16 GB of VRAM that I had just lying around because I saw it on sale for a couple hundred bucks and I needed a temporary GPU replacement for a smaller PC. I’ve tried running LLMs on it before and it’s not… super fast, but it’ll do what you want for 14B models just fine. That’s going to be your sweet spot on home GPUs anyway, anything larger than 16GB and you’re talking 3090, 4090 or 5090, pretty much exclusively.

This is… mostly right, but I have to say, macs with 16 gigs of shared memory aren’t all that, you can get many other alternatives with similar memory distributions, although not as fast.

A bunch of vendors are starting to lean on this by providing small, weaker PCs with a BIG cache of shared RAM. That new Framework desktop with an AMD APU specs up to 128 GB of shared memory, while the mac minis everybody is hyping up for this cap at 24 GB instead.

I’d strongly recommend starting with a mid-sized GPU on a desktop PC. Intel ships the A770 with 16GB of RAM and the B580 with 12 and they’re both dirt cheap. You can still get a 3060 with 12 GB for similar prices, too. I’m not sure how they benchmark relative to each other on LLM tasks, but I’m sure one can look it up. Cheap as the entry level mac mini is, all of those are cheaper if you already have a PC up and running, and the total amount of dedicated RAM you get is very comparable.

I don’t know how you fix that problem, but I’ll admit that you do need some functional anti-cheat. Nobody wants to go back to the days of PC gaming being the wild west while consoles were nice and secure.

I wonder if in a world less focused on Windows some multiplayer games would just work on some secure container type of thing, or just have most of the gameplay run on server or something. There are definitely other solutions that wouldn’t rely on the Windows-specific crutches of the current implementations.

I heard they finally have official support. How well it works I don’t know. I haven’t tested it yet.

I think the assumption that people are going to have AMD hardware is a bit of an issue with this argument. Even with their current gen success they are under 10% of the market. That’s all good for committed Linux users who built their PCs with Linux compatibility in mind, but 90% of the desktop market (for gaming at least) is going to be repurposing a Windows device with a dedicated Nvidia GPU.

You’d think, but at least in my Manjaro install I had the exact same, if not a bit worse, of an experience trying to share an exFAT drive than a NTFS drive. I don’t recommend it either way.

I definitely play enough games without full Linux support that I wouldn’t have switched fully, even if I didn’t need Windows for work. The anticheat issues are one thing, but with a high end Nvidia card I found a bunch of proprietary features either didn’t work or underperformed compared to Windows. Mix that with a HDR, VRR display and it was a bit of a mess.

Linux was snappier for desktop office work most of the time, though.

Hosting the games on NTFS and loading them into Steam from there under Linux is possible. It is inconsistent and a hasssle, though.

I will say the setup the OP suggests is totally doable, but when I’ve had it that way it turned out to be easier to just do everything else on Windows than to flip back and forth, so after I updated some hardware I haven’t been on a hurry to set up Linux again.

I’d say it’s more convenient to do this long term if you have two PCs. Maybe a laptop for Linux work and a desktop with a powerful GPU for gaming. Being able to have both on sleep and quickly switching back and forth is less likely to make you (well, me, at least) lazy than having to reboot each time.

Who cares? I’m confused. Why is their upload relevant in the first place? I thought all the IP holders were out there arguing that the download was the issue.

Never mind that in this case there is a profit reason for the download in the first place. Even in notoriously lenient areas with copyright that is a bigger strike than whether they reseeded anything in a peer to peer platform.

But hey, here we are, I’m rooting for Meta here. Absolutely put all the money in dismantling overreaching copyright regulation. Let’s find some Duck Tales comics or whatever in there. Disney vs Meta in court over copyrights. The Godzilla vs Kong our generation deserves. Let’s do it.