Doesn’t matter; they’re never going to be profitable. That’s not the direction the field is headed.

So we can deal with it now. Or later, when things are worse.

Absolutely fucking not.

I’m not coming from the “AI hater” angle either. I love open models.

But OpenAI and Anthropic are abominations. I’ve watched good startups and research groups disappear into the gaping maw they’ve created, all reason go out the window, replaced by whatever lies are coming out of Altman and Modi’s mouths.

…They’re destroying US innovation, and sucking everything into a black hole.

They need to burn.

And its honestly insane someone like Schneier can’t see this, or was just paid to say this. No one in the machine learning research space likes OpenAI; they never have. I guess he’s old enough to use past nationalization efforts as a baseline, in his mind?

OP, can you link the original research? It’s an excellent site:

https://weloveit.io/developers/steam-revenue-report/

What’s even creepier is that Facebook would track the snot out of anyone interacting with the ad. It would aggressively serve them more content like that.

…Don’t people have any shame, even if they’re into weird shit? What they’ve been tapping will be obvious to anyone who looks at their feed.

Nah. This is cheap porn to sell to the lowest common denominator, that’s all.

Nudes classification predates LLMs by decades. Its accurate, and dirt cheap, because it had to be to detect CP.

You can bet Lemmy.world uses it. Plenty of smaller companies do it. Pick any NSFW picture on CivitAI, and you can see how insanely accurate autotagging has gotten.

So, no… this was not an accident on Facebook’s part.

YMMV.

Big ZRAM absolutely tanks performance for me because it “steals” space that could be used for normal memory.

But if its for squirrling away a bunch of idle processes when multitasking like crazy, yeah, it can be wonderful. Just depends.

Depends what you are going for. 7000/9000 series have way more per-thread performance than threadripper.

Another quirk with single CCD 7000 chips (which I have as well) is that the infinity fabric limits effective bandwidth to the CPU cores. Even with two channels, you hit diminishing returns with higher speeds, especially if you aren’t using the IGP.

You can get around it overclocking the IF, but it gets complicated and its much easier to do with 9000 series CPUs.

I do.

I wish I had 192GB, to be honest.

So I got this RAM in August 2025, for ~$400 according to the receipt:

https://www.newegg.com/g-skill-flare-x5-128gb-2-x-64gb-ddr5-6000-pc5-48000-cas-latency-cl36-desktop-memory-black/p/N82E16820374767

…That is absolutely bonkers.

The Feng Shui is strong, with this one.

LLMs (especially Gemini) have a tendency to hallucinate things from extant references, especially with a huge blob of spammy search results in context.

Even with a real reference, you have to be careful.

…TBH, I’ve been fighting fire with fire. Sometimes I get local Qwen 27B to point to the relevant bit in a reference. If there is one.

I hope that we’ll collectively decide that it’s rude to pass off LLM writing as your own, but I wouldn’t bet on it.

We won’t unless we stop the source perpetuating this “culture.”

Big Tech.

Unfortunately, they kinda have a grip on everyone’s information sphere, so I think our chances are low :(

Yeah.

It’s crazy to think they wouldn’t, like with any other form of plagiarism or fraud. AI isn’t some special case that somehow gets people out of responsibility.

He says Kimi is one big girl, and Qwen is soooo thin.

This is true of all OSes TBH. Issues arise when you start to mess with them.

Remember this if you switch to linux.

Even “big” open source models like DSV4 and Ling/Ring are very efficient. They’re big, but (seemingly) sparser than US models, so they’re cheap.

They run surprisingly well with hybrid CPU+GPU inference on desktops. And thats not even getting into the efficient attention mechanisms.

I can run DSV4 Flash, barely quantized, with ~1M context on my Ryzen desktop at ~11 tokens/s. If you told me that two years ago, I would not have believed you.

MS’s antics extend well beyond Windows, unfortunately. There are long blogs and writeups dedicated to it.

But what you said is also kinda true. 7 and XP worked reasonably well as long as you didn’t mess with them, push their bounds too much, or use them for anything “weird.” But you also probably missed a lot of dysfunction in the background.

I’m not trying to shill for Windows 7/XP era OSX or linux or anything either. But Windows has always been kinda notorious.

Like Scipitie said, sort of.

In the ChatGPT app, there’s tons of shuffling in the background, like context being injected, maybe sampling changed, agenic action, quantization… its “far” from the actual LLM, opaque by design, and as a result certainly not deterministic.

You can sometimes get deterministic output with the OpenAI API, but it’s also dependent on nothing changing on their end. And their end changes a lot.

But self hosting or using a more consistent provider will give you deterministic output.

We’re still talking about Microsoft, right?

They have a looooong history of enshittification and footgunning that predates “AI” by decades, no matter how much the world has depended on their stuff. So no, I suppose I wouldn’t think that, as that would break decades of precedence for MS. But I suppose I never trusted them in the first place.