llamasushi

Just updated, thanks

The burying of the lede here is insane. $5/$25 per MTok is a 3x price drop from Opus 4. At that price point, Opus stops being "the model you use for important things" and becomes actually viable for production workloads.

Also notable: they're claiming SOTA prompt injection resistance. The industry has largely given up on solving this problem through training alone, so if the numbers in the system card hold up under adversarial testing, that's legitimately significant for anyone deploying agents with tool access.

The "most aligned model" framing is doing a lot of heavy lifting though. Would love to see third-party red team results.

This is amazing. Thank you for this.

But does it work on GOODY2? https://www.goody2.ai/

That's an optimistic take, but equally valid is the take where delusions provide impetus for some pretty nasty behaviour - eg see the crusades

"Warmer and more conversational" - they're basically admitting GPT-5 was too robotic. The real tell here is splitting into Instant vs Thinking models explicitly. They've given up on the unified model dream and are now routing queries like everyone else (Anthropic's been doing this, Google's Gemini too).

Calling it "GPT-5.1 Thinking" instead of o3-mini or whatever is interesting branding. They're trying to make reasoning models feel less like a separate product line and more like a mode. Smart move if they can actually make the router intelligent enough to know when to use it without explicit prompting.

Still waiting for them to fix the real issue: the model's pathological need to apologize for everything and hedge every statement lol.

LeCun, who's been saying LLMs are a dead end for years, is finally putting his money where his mouth is. Watch for LeCun to raise an absolutely massive VC round.

Lol, this reminds me of a funny story. I had a lawyer whose name was Jim Halpert. Turns out he was the very Jim who was inspired his namesake on the office. Asked him about it once. His reply? "Hey, it's been great for getting clients." =)

He was also very much like Jim on the show. Fun times.

Am I the only one who thought this was referring to how people felt about the general zeitgeist? Like, how Romans viewed everyone outside Rome as barbarian, etc. Not in the literal sense like, mirrors. Nice HN switcheroo.

Yeah, this was my first startup and I thought I could multi-task lol. Probably did it for the wrong reason (eg: "checking a box - went hitchhiking in Alaska") than anything else. Spent most of the trip worrying about other) startup shit, didn't enjoy the scenery nearly as much as I should have. Still regret it haha!

So it's not really hallucinating - it correctly represents "seahorse emoji" internally, but that concept has no corresponding token. lm_head just picks the closest thing and the model doesn't realize until too late.

Explains why RL helps. Base models never see their own outputs so they can't learn "this concept exists but I can't actually say it."

Went hitchhiking in Alaska while running my startup to "get a break". Absolute disaster. Couldn't properly connect to internet, dropped a bunch of meetings, etc.

Still worth it. My fault for not planning in advance.

You'd be surprised. America is an incredibly fucked up society. Free meals, Healthcare. Hell, even cheaper tattoos. There is a certain segment of the population where "jail" does not have the same meaning as it does for you or I.

Small anecdote, but back in early 2024 (like, March), I did a ton of searching for an AI helper app that would allow me to use the gemini API key from AI studio. Chatbox was one of only two that I found (can't remember the other), but even back then I was astounded by its functionality and ease of use. It supported a ton of stuff, like custom system prompts, etc. Basically recreated a lot of the lmstudio experience but on a phone.

You'd be surprised, but it was extremely difficult to find an android app that supported API key usage. Ahead of its time.

Lol, tether, bitfinex are examples that came to mind. A lot of the OG crypto instutions got to where they are by "faking it till they made it" long enough to actually make it.

Does no one still remember that tether continually stalled audits FOR YEARS in the face of increasing scrutiny?

One doesn't need to go more than 2 feet into the mire of meme coins before finding the detritus of 6000000 rug pulls. Just that these guys never get prosecuted.

The compute moat is getting absolutely insane. We're basically at the point where you need a small country's GDP just to stay in the game for one more generation of models.

What gets me is that this isn't even a software moat anymore - it's literally just whoever can get their hands on enough GPUs and power infrastructure. TSMC and the power companies are the real kingmakers here. You can have all the talent in the world but if you can't get 100k H100s and a dedicated power plant, you're out.

Wonder how much of this $13B is just prepaying for compute vs actual opex. If it's mostly compute, we're watching something weird happen - like the privatization of Manhattan Project-scale infrastructure. Except instead of enriching uranium we're computing gradient descents lol

The wildest part is we might look back at this as cheap. GPT-4 training was what, $100M? GPT-5/Opus-4 class probably $1B+? At this rate GPT-7 will need its own sovereign wealth fund

Yeah, this one line gave it away for me: "you’re not just writing contemporary code—you’re building applications that are more maintainable..."

Look up dialectical hedging. Dead AI giveaway.

Piggybacking off this, for a more general reason for sleep: "My definition would be as follows: sleep evolved as a species-specific response to a 24-hour world. During sleep – a period of physical inactivity – individuals avoid movement within an environment to which they are poorly adapted, but then use this time to undertake essential housekeeping functions demanded by their physiology."

From Life Time by Russell Foster. Still one of the most lucid and well-written books on sleep I've ever read.

https://archive.is/20250730074231/https://www.bostonglobe.co...

Whole trip felt more like what i would imagine visit to mainland China would be like rather than a trip to a free western country.

Have you ever been to mainland China? I've lived in both places and honestly, day-to-day life in major Chinese cities often feels more "free" in practical ways - safer, cleaner, more technologically convenient.

What is freedom really? In Shanghai or Shenzhen, I can walk out at 3am to get noodles or take the metro without a second thought. In LA or SF, I'm constantly aware of my surroundings, checking who's behind me, avoiding certain areas. The surveillance cameras in China never made me feel as watched as the constant threat assessment you do in many Western cities.

Don't get me wrong - I'm not saying China doesn't have serious issues with political freedoms and surveillance. It absolutely does. But the lived experience is way more nuanced than "oppressive dystopia."

I used to have similar assumptions before actually spending time there. Western media coverage (cough propaganda cough) tends to focus exclusively on the authoritarian aspects while ignoring that for many people, daily life feels safe, convenient, and yes - "free" in ways that matter to them.

Instead of imagining what China might be like based on western news coverage, why not visit and see for yourself?

Extremely unpopular opinion on HN, I'm sure. But I have a compulsion to challenge stereotypes when the reality is so much more complex.

I appreciate this comment. You will see that the modern day capitalistic system, in general, punishes anyone with even a smidgen of the moral compass you have. The world of finance is this in spades. Having worked on wall street for pretty much my entire adult career, having gone on to found my own fund. I came to an epiphany through a few fucked up experiences that my investors did not give two flying fucks what kind of person I was as long as I was generating solid returns. Moral compass be damned.

So, casino industry perhaps is a convenient pinata when in reality it's not the specific industry, it's the system.

Why the down votes here? This is an interesting idea.

Socrates: "And now, since you are the father of writing, your affection for it has made you describe its effects as the opposite of what they really are. In fact, it will introduce forgetfulness into the soul of those who learn it: they will not practice using their memory because they will put their trust in writing, which is external and depends on signs that belong to others, instead of trying to remember from the inside, completely on their own. You have not discovered a potion for remembering, but for reminding; you provide your students with the appearance of wisdom, not with its reality. Your invention will enable them to hear many things without being properly taught, and they will imagine that they have come to know much while for the most part they will know nothing. And they will be difficult to get along with, since they will merely appear to be wise instead of really being so."