Probably, but China is REALLY good at scaling manufacturing, in general. Hopefully they’ll be able to quickly scale up their memory production.

Why should that matter to anyone other than the owner of the PC? There are probably some games that I’d consider a “stupid waste of time,” but I couldn’t care less what games people play on their own PCs. That’s nobody’s business but their own.

(Side note [not that this matters]: Gaming often consumes more power than LLM inference.)

Yes. Chinese labs keep improving small, open-weight models. IIRC, Qwen3.8-27B benchmarks are comparable to frontier models that were released less than a year ago.

I run that model on my RTX 5060 Ti 16GB using ExLlamaV3. Though, I’m not sure if that qualifies as “standard PC” hardware as it now costs more than 2x the price I paid a year ago :/

How much power do you think local inference actually consumes?

If I had to take a wild guess: Government bailouts

Yeah probably a lot. And I agree. My first sentence covers those too. The rest of my comment was just LLM-focused because that was a little more on-topic. But yeah, I don’t feel a sense of privacy with any of those kinds of cloud services, AI or not.

It’s always interesting to see how much people trust sending such personal information to some service on the Internet, expecting privacy.

I’m not sure I’d even want my own self-hosted LLMs to have something like a personal diary. Sending all that to an AI company would never even cross my mind.

I understand that most people aren’t the tech-oriented types who get into self-hosting and running their own LLMs, but people have had diaries for centuries before LLMs existed.

Care to elaborate? (Or point to where one can learn more about this?)

How would that help?

You want guys to just leak piss onto whatever surface they’re standing/sitting on?

Here’s a practical use case: Sending money internationally.

I wanted to send some money to a friend in another country once, and it was difficult to find some service that could do it without some crazy expensive fee.

I’ve sent money to other countries before, but for this particular country, it was different. Maybe it’s some (dis)agreements between our governments, banks, etc. Maybe it would require the service to send it through another country or two, or convert it to some other intermediate currencies. I didn’t look into it, so I don’t really know why it was so expensive.

Anyway, it was much simpler and cheaper to just send crypto instead.

Yeah that was the idea, originally (hence why they’re called cryptocurrencies). It’s unfortunate that that’s not what it’s actually used for, in most cases.

There were also environmental concerns, but AI seems to have taken that spotlight

Yep. It’s also interesting to see teams have conversations about implementing these things without anyone mentioning how creepy it seems. I guess these things are just normalized now :/

Yeahhh that sucks that we’ve reached this point. I would actually consider paying some sites for my privacy. If they offer something that I value, then it’s fair for them to be compensated somehow. However, if they offered that option, I still probably wouldn’t trust it.

Even if the site itself fully intends to stand by their offer, can I trust the payment processor(s), bank(s), etc. involved in the transaction?

I’m not really a crypto guy, but I wish something like Monero (or whatever privacy-protecting alternative that might be “better” these days) could become more mainstream.

I have not. Is that game only about torturing the characters, or is that more of a side quest?

🤔 I don’t think it’s quite the same as video games. Video games usually involve some challenge, goal, risk of losing, etc.

There are only a few games that I play though, so I might not be “qualified” to say for sure, but a torture simulator with no purpose outside of torture doesn’t seem like much of a game.

To spend this amount of time and effort building the necessary software environment for “torturing” LLMs is creepy. Like, imagine an adult who builds a bunch of crazy contraptions to physically “torture” some dolls or something. It harms no one, but it’s creepy.

Heh, guess Gemini wasn’t even worth including

My Hermes instance periodically checks the overall open-weight LLM landscape, and it recently recommended trying Ornith 1.5 35B-A3B.

I haven’t tried it yet, but on paper, it sounds like it has some potential.

🤔 what happens if/when we reach ASI? Do we just call it Super-Duper Intelligence?