Yeah it’s nuts. I have a small bladder and just naturally pee a lot. I usually went to the toilet between every lesson at school. Is moving schools an option for your kids?

I’m sure you’re probably right. I haven’t looked into how things like clock speed/FLOPS/etc affect LLM performance.

The main thing is you need ROCm/CUDA/Vulkan support. If you do then just try running some models that fit into your VRAM and see how they perform? On my card, it takes maybe a minute or two to generate an answer, sometimes several minutes if the answer is long. I use small models that are <8GB in size, so not the best open-weight ones, but I don’t use LLMs for anything particularly complex anyway so the dumb ones do the trick for me.

Radeon RX 7600, but yes, you’re right, it’s relative. It’s good enough for any gaming I do, which is not an awful lot. I think it’s considered mid-range for its generation, I guess I was just thinking of the GPUs I see people mention they have which seem much beefier than mine lol.

I’m no AI expert but when I looked into it, it seemed that VRAM was the biggest factor being mentioned in discussions of GPUs, hence why I mentioned it.

Not much, but you can use it for basically anything people use frontier models for, it’ll just be somewhat dumber. I occasionally run LLMs for proof-reading important things that I’ve already hand-written and edited. Sometimes for bouncing off ideas. Sometimes for the kinds of questions I’d normally chuck at a forum but I can get a quicker answer if I ask an LLM. No, I don’t “trust” the answers; anything where I can’t just use my own judgement, the LLM will have given me enough to perform my own research to fact-check it.

I have a lower-end GPU (8GB VRAM) and I can still run smaller models on it with decent performance.

IVPN or AirVPN (depending on whether or not you need port-forwarding) is my VPN recommendation.

It would likely have to have a large pool of accounts and randomly pick one per request, to avoid Reddit noticing a too-high volume of requests from one account.

IVPN has been working well for me. Heard AirVPN is good if you need port forwarding too.

Reuters blocks my VPN and the Wayback archived version doesn’t display anything.

They, notably, did not say that PHP was good.

Literally just having ollama installed and some models that work well with the amount of VRAM I have. I don’t auto-start the daemon because I rarely use it. Occasionally I use it for proof-reading/feedback on things like cover letters, or debugging. I don’t let it write anything. If it has something useful to say it can tell me in the chat and I can implement it if it’s good. Usually I don’t need it though so it just sits on my disk for the most part.

Wow, this article really reads as LLM-generated to me. The quoted section with the “We won’t pretend otherwise:”, the second paragraph in:

This post isn’t about how to update. It’s about how these vulnerabilities were found and fixed, what has changed in how PHP security is handled over the past two years, and what we need from the organizations that depend on PHP to keep PHP’s security work going.

Could they not be bothered to write a write-up about such a critical severity?

A lot of FOSS solutions have funny names, mascots, etc. I’m surprised if IT/tech workers aren’t used to it. Just politely tell them, yes, it has slightly silly branding, but look past it and it’s software that does what you want.

Yes, it can be a bit involved. I’d say I can spin up a mail server in 1–2 days though provided the OS is already installed and I have a domain name and port 25 isn’t blocked (that’s the main one that tends to be blocked). It may take longer for people who aren’t familiar with mail servers but it’s not some huge operation.

You can just try spinning up a mail server for fun. If it goes badly then just don’t move over your dentist appointments to it. It’s only critical infra if you use it for critical things, which you probably shouldn’t till it’s stable.

I self-host my own email, have done so for years. I’ve not gotten pwned. Use an SSH key and a firewall, only expose the ports that should be public, have reasonable access controls, etc, just practise general cybersecurity—it applies to servers too.

I know everyone complains about hosting email. But I’ve not run into major problems. My emails get through spam filters and I have good uptime. It’s also a fun learning experience.

A small point but it assumes that your browser’s self-declared timezone is “true” to where you live. My timezone is spoofed to UTC+0; my VPN server is in a different country; and my browser language is set to en-us. I feel like if you see a user whose settings are all of the above then you can assume that none of those three data points actually describe the user, unless they coincidentally are a US English speaker or live in UTC+0, but that’d just be them coincidentally living where anti-fingerprinting browsers report you as.