RNG is not an inherent property of a transformer model. You can make it deterministic if you really want to.
You can’t convert it back into anything remotely resembling human-readable text without inference and a whole lot of matrix multiplication.
Could you not make a similar argument about a zip file or any other compression format?
:x does the same thing as :wq but in one less keystroke :3
I think that, while yes, LLMs are an option for data storage, I don’t think that they’re worth the effort. Sure, they might have a very wide breadth of information that would be hard to gather manually, but how can you be sure that the information you’re getting is a good replica of the source, or that the source that it was trained on was good in the first place? A piece of information could come from either 4chan or Wikipedia, and unless you had the sources yourself to confirm (in which case, why use the LLM as all), you’d have no way of telling which it came from.
Aside from that, just getting the information out of it would be a challenge, at least for the hardware of today and the near future. Running a model large enough to have a useful amount of world knowledge requires a some pretty substantial hardware if you want any amount of speed that would be useful, and with rising hardware costs, that might not be possible for most people even years from now. Even with the software, if something with your hardware goes wrong, it might be difficult to get inference engines working on newer, unsupported hardware and drivers.
So sure, maybe as an afterthought if you happen to have some extra space on your drives and oodles of spare RAM, but I doubt that it’d be worth thinking that much about.
I doubt they’re blaming the site, they just lost it and are trying to find it again.
I’m in the latter camp mostly because on top of the inconsistencies in the images, (Why are there two dishwashers? Why is the dog climbing into the dishwasher instead of reversing into it as described?) they don’t really have a plausible reason to exist at all, at least in the context of the text. Why is the dog holding the box and looking directly at the camera in the first photo? If the dog really was flailing around like was described, how would the person have time to take not one but two perfectly framed and posed shots?
oalouub is my favorite cheez-it flavour!
The point the tweet OP is making isn’t that they want Musk to be deported to South Africa, they’re just calling out his hypocrisy.
Your websites have updates??
OOTL, what’s up with NZXT?
Oh, my bad! The wording didn’t parse as humor to me.
It was a joke, friend
Bot account? Comments seem like your average “short and humorous response” bot.
Figures
Who’s the artist of this comic?
It’s not the LLM that does the web searching, but the software stack around it. On its own, an LLM is just a text completer. What you’d need a frontend like OpenWebUI or Perplexica that would ask the LLM for, say five internet search queries that could return useful information for the prompt, throw those queries into SearxNG, and then pipe the results into the LLM’s context for it to be used.
As for the models themselves, any decently-sized one that was released fairly recently would work. If you’re looking specifically for open-source rather than open-weight models (meaning that the training data and methodologies were also released rather than just the model weights), GPT-OSS 20B/120B and the OLMo models are recent standouts there. If not, the Qwen3 series are pretty good. (There are other good models out there, this is just what I remember off the top of my head.)
Qwen3-0.6B is about 400 MB at Q4 and is surprisingly coherent for what it is.
I have to ask, why buy a key for the VM at all? Windows functions perfectly fine without one, and you can always use MAS if you want to change your wallpaper or something.
Games should have a point, and winning is not a point on its own.
Why not? Is wanting to win not a valid motivator to play a game?
Huh, I didn’t know that. Neat!