Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I love LLMs, I hate hype

geohot.github.io

Post reply on HN
Print view View original

161–170 of 340 posts

Re: I love LLMs, I hate hype

#161

Post by vardalab »

anuramat wrote: post #157
maxgashkov wrote: post #144

Earlier quoted context omitted.

If they haven't relaxed the classifier this makes little to no difference as the model is essentially unusable no matter the token quotas.

what are you working on? I only hit the guardrails twice after burning through two weeks of 20x max plan, both times on ML stuff; still more than I'd want to, but not unusable

A lot of stuff that has to do with VLLM and troubleshooting and compiling and building VLLM, compiling kernel, or just dealing with setting up eval for local models, it punts to Opus 4.8 on a regular basis. To the point that I have given up on using it for that purpose.

Re: I love LLMs, I hate hype

#162

Post by stavros »

seattle_spring wrote: post #154
stavros wrote: post #150

Earlier quoted context omitted.

I know that some companies are spending around that much per day per engineer . They just want to go as fast as possible.

They're spending $10 million USD per engineer on LLMs?

Sorry, I meant per month, not per day.

Re: I love LLMs, I hate hype

#163

Post by fragmede »

deep_concern wrote: post #69

>One, this constant bullshit about some window closing, or the perpetual underclass, or falling hopelessly behind. This is negative valence hype, not only is it not true, it’s mostly designed to make you feel bad about yourself and move to shitty San Francisco where everything really does suck like how these people claim. It's bullshit in the sense that they don't know for sure, but the author doesn't either. Why mig…

No one knows for sure. I certainly don't. Looking at history though, at what happened in 2008, and the effects it had on my own personal financial situation, it's easy to see "falling behind" as plausible.

Re: I love LLMs, I hate hype

#164

Post by fragmede »

andy99 wrote: post #121
SwellJoe wrote: post #81

This line: "this is my main argument against the valuation of frontier labs. It’s not that AI won’t create that much value, it’s that they won’t capture it." That is a very astute and concise way to explain everything about how the frontier labs are behaving and how they're trying to push more people to pay token rates for the best models. At the current subscription prices ($100 or $200 a month for a generous, thoug…

The funny thing about Fable, is we all but know it will be obsolete in a month or two. Between their embargo shenanigans (which IMO they could have avoided just by not pretending it was dangerous) and continuing to give access, whatever marginal advantage it had was essentially wasted. It would have been an interesting experiment to charge more for it right away and see what the market would bear, rather than tease i…

Mythos was first publicly known via a leak at the end of March, 2026. Given that Fable is the public version of Mythos, six months gets us September for when the next big leap will happen.

Re: I love LLMs, I hate hype

#165

Post by bluefirebrand »

bluegatty wrote: post #158
SwellJoe wrote: post #143

Earlier quoted context omitted.

But, that's not what you said they'd do. I can switch to a different model with almost zero cost. That's the definition of a commodity.

No, you switched to another SOTA model. You didn't switch to 'Random Corner Store Token Seller' down the street, did you? There 2-3 top players, that is not commodity. Commodity is when there are enough that none of them have market power or can set prices. 'Commodity' means you buy your tokens from the Grocery Store on their loan plan. Like consumer credit is a commodity. FYI predict this is roughly the way it will…

> No, you switched to another SOTA model. You didn't switch to 'Random Corner Store Token Seller' down the street, did you?

You're assuming that SOTA never hits a hard ceiling, letting local models catch up and achieve parity

It seems unlike that the frontier labs are going to be keeping ahead forever, they'll hit some kind of ceiling eventually

Re: I love LLMs, I hate hype

#166

Post by fragmede »

roncesvalles wrote: post #149
SwellJoe wrote: post #81

This line: "this is my main argument against the valuation of frontier labs. It’s not that AI won’t create that much value, it’s that they won’t capture it." That is a very astute and concise way to explain everything about how the frontier labs are behaving and how they're trying to push more people to pay token rates for the best models. At the current subscription prices ($100 or $200 a month for a generous, thoug…

>5.6 Sol, competitive with Fable (or close enough), and it's available via subscription (even the $20 subscription!) It's not comparable because OpenAI caps thinking to High in the ChatGPT "Chat" interface (and the "Work" thing where it actually does let us use Extra or Max is fucking shit). GPT 5.6 Sol (High) is almost certainly worse than Opus 4.8 (Extra), and nowhere close to Fable (Extra). I literally got a refun…

5.6 Sol isn't Fable, but calling it unusable is a bit much.

Re: I love LLMs, I hate hype

#167

Post by wccrawford »

lacedeconstruct wrote: post #122
dolebirchwood wrote: post #118

Earlier quoted context omitted.

> This doesn't make sense Makes perfect sense to anyone good at using these models. What doesn't make sense is that analogy. Typing prompts isn't even close to as difficult to baking bread.

>Makes perfect sense to anyone good at using these models. It doesn't really, because whenever I ask them what did they actually create, its always a shitty dashboard or a finance tracker or something derivative and worse than what is out there

You're missing the point that it isn't worse than what's already out there for them.

They've implemented only the parts they need, and removed all the crap that just complicates the system for them. They've made it do exactly what they need, exactly how they need it. It's something that you couldn't afford to do previously, sometimes even as a programmer. Now, it's often quick and easy, up to a certain complexity.

Re: I love LLMs, I hate hype

#168

Post by maxgashkov »

anuramat wrote: post #157
maxgashkov wrote: post #144

Earlier quoted context omitted.

If they haven't relaxed the classifier this makes little to no difference as the model is essentially unusable no matter the token quotas.

what are you working on? I only hit the guardrails twice after burning through two weeks of 20x max plan, both times on ML stuff; still more than I'd want to, but not unusable

very mundane web app and networking security (vpn etc.), nowhere near redteaming

Re: I love LLMs, I hate hype

#170

Post by mistrial9 »

vatsachak wrote: post #47
apsurd wrote: post #16

Earlier quoted context omitted.

damn, you all hate SF that much?

I don't hate SF it's just overpriced. Whenever I visit SFO it's really funny seeing all the advertisements from startups above a population struggling to find housing. Won't it be better to pay someone 100k in Reno than 180k in SF? Most collaboration happens online these days anyways. Honestly 60k in Barcelona is like 200k in SF when you look at housing and public services. We need to punish bad city governance for b…

San Francisco has always been a crooked city

source: Barbary Coast USA

Post reply on HN