HNHacker News
TopNewBestAskShowJobs

qeternity

8,841 karma · joined May 4, 2016

submissionscomments
qeternity··on Mistral Large 4
As an American living in Europe, it's generally Europeans who have a distorted perception. Most Americans know the bill of rights and the 1st/4th/5th amendments. Many Americans will know of the Patriot Act/NSA/Snowden revelations.

The same is not true of Europeans and they generally believe their rights are much stronger than they are. Most Brits are not aware of the true nature of the Investigatory Powers Act. Most Frenchmen are not aware of Article L851-3.

qeternity··on Mistral Large 4
Do you understand what "metaphorically speaking" means?
qeternity··on Mistral Large 4
> Generally speaking, the EU has some quite strong privacy protections relative to the US.

Please, like what? GDPR?

qeternity··on Mistral Large 4
> PATRIOT, FISA and Bush's surveillance programme imho give more powers to certain agencies today already than are codified in EU law.

This is not true. Whether you believe that FISA courts have teeth, or whether 3 letter agencies abide by the law, in most of Europe you do not even have the pretense of this. UK/FR and to a lesser extent DE are all much worse.

The worst parts of Patriot Act were undone in 2015, and the "F" in FISA stands for "Foreign". Europe continues to push policies like Chat Control and attempt to backdoor/ban E2E encryption for domestic surveillance.

The present day and trajectory in Europe is far far more grim than in the US.

qeternity··on I quit OpenAI because its culture is broken
I'm not advocating anything, to be clear.
qeternity··on Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s
> Flash attention with who knows how much draft ensures it makes the same mistakes all the time and cannot call tools, format output or follow basic guidelines reliably.

> Draft MTP <= 2 so it doesn't trip

I am not sure you understand what either of these things do.

Do you think that FA or MTP are lossy?

qeternity··on I quit OpenAI because its culture is broken
So does manufacturing GPUs.
qeternity··on When oil prices spike, where does the money go?
That trade works when the market is in steep carry, usually a result of depressed prices.

That trade does not work today when the market is very inverted: you buy the spot oil, and sell a call in the future. But the term structure is pricing lower oil prices in the future. You take that hit on your spot cargo.

That trade only works if you are bearish the curve and bullish flat price. And if you have this view, it's just about the worst possible way to structure that trade.

The cure for high prices is high prices. Curve inversion ensures that anyone with oil today is incentivized to sell it asap.

qeternity··on Ember-1
This is undoubtedly great. But most of the inference cost today for dominant use cases (agentic coding) are in the prefill, not the decode. This is one of the reasons that DeepSeek is so aggressively optimizing prefill and caching.
qeternity··on Show HN: Trader News – Hacker News for Finance
Hugged to death?
qeternity··on OpenJev
Not sure why you think everyone has moved on from this. Valid JSON is only one aspect. It almost certainly has far more uptake today than ca. Sonnet 3.7. Grammar engines can enforce a lot of other things too. Models are actually really good at emitting valid JSON, but complex schemas effectively require it.
qeternity··on OpenJev
Alternative is to fail the validation and regenerate the request, but I have no idea what this person is talking about. It is absolutely not the case that everyone has moved on from structured generation. Absurd claim.
qeternity··on DeepSeek launching v4.1 flash cheaper and more capable than v4 pro
Most people are not even referring to CUDA batching nuances.

They think that sampling is an inherent part of Transformers.

Even on this site, it is regurgitated with confidence.

qeternity··on Discovery of a new OpenAI agent message board
It’s only good for them in the sense that it is allowing them additional time and compute to cheat a reward signal.

It is not good for them in the sense that they will short circuit the RL path that actually improves general capability.

qeternity··on Apple caught off guard by AI demand for Mac Mini and Mac Studio
Has nothing to do with being headless. That's just a natural outcome.

If someone wanted to use an 8x B300 as their daily driver...go ahead.

It would still be the best way to serve a given model.

qeternity··on A practical guide to running 8x RTX PRO 6000's
*TB/s
qeternity··on GLM-5.3-Flash
https://typebulb.com/u/lab/you-re-relatively-right/full
qeternity··on Apple caught off guard by AI demand for Mac Mini and Mac Studio
> Edit: I'm thinking of a headless Mac mini, if you meant running it on the same machine you're using of course you'll need more memory, but LLMs are best served from a headless server so that's what I'd recommend.

What? LLMs are best served from a massive PD disaggregated cluster of B300s connected via NVLink.

If you're running LLMs on a Mac Mini, it's because you want to run local, not because it's the best setup.

qeternity··on Apple caught off guard by AI demand for Mac Mini and Mac Studio
You're not entitled to this stuff any more than they are.

What planet do you live on?

qeternity··on GLM-5.3-Flash
Trolling. GLM is heavily distilled from Gemini.
qeternity··on Coconut oil jet fuel matches kerosene's efficiency in engine tests
Is it not a bad thing to encourage poor people to continue doing the thing that has apparently made them poor?

It's also very not true that farmers are poor in the US (median household income about 30% greater than general population). This mindset is like a century old.

qeternity··on AI boosted homework scores, then exam scores dropped: study
College stopped being about getting an education a long time ago.

It's just a low pass filter for the job market. Students aren't actually interested in learning anything, they have been condition for a long time to jump through the hoops.

It would be much better if we actually returned to exam and performance based evaluation of students, and stopped giving everyone As just for trying hard. But parents look at college fees as buying their kid a job opportunity, and so grade inflation has basically ruined the SNR of academic performance for all but the lowest performers.

qeternity··on Felony charges for citizen deleting phone data at US Border
It's the other way around: conviction rate includes plea bargains.

Convictions at trial are much lower.

qeternity··on Felony charges for citizen deleting phone data at US Border
You are misunderstanding the conviction rate. Only ~2% of cases actually go to trial, and at trial there is ~80% conviction rate.

There are ~4x as many cases dismissed by judges before getting to trial. And the vast majority (90%) of defendants enter into plea bargains.

Prosecutors only bring charges when they feel they have a strong case. There are many many cases which are never pursued because of this, and people also get upset about that.

Many countries do not have a plea bargain system the way the US does. And if you look at conviction rates at trial, they are smack in line with much of e.g. Western Europe.

qeternity··on What Happens When the Cost of Intelligence Drops 100x
I presume you're a Windows user then.

When the M1 was released, I couldn't believe how fast it was.

qeternity··on Cerebras CS-4
Ultimately the frontier labs are competitors of Nvidia. There is a fixed amount that the market will pay for tokens. If the frontier lab model premium collapses due open weights models, Nvidia can capture a greater share of aggregate spend.
qeternity··on U.S. Debt Hits $40T as America's Borrowing Binge Continues
It is still people, not money, who votes for them over and over again…
qeternity··on Qwen 3.8 27B is excellent, but it defaults to overthinking things
Personal hardware is only sufficient today for some tasks. As data center power efficiency, and large sparse MoE task efficiency increase, personal computing will continue to lose out.
qeternity··on Qwen 3.8 27B is excellent, but it defaults to overthinking things
Computer use will blow through tokens because it's doing image capture for everything.

You may have better and more reproducible results using browser controls that aren't image based, or writing tools that completely sidestep browser use.

qeternity··on Qwen 3.8 27B is excellent, but it defaults to overthinking things
Jevons paradox: large purpose-fit data centers increase efficiency such that you can use AI in more places, and use more tokens for those tasks.

The future is not a single chat bot session of bs=1. The future is many agents performing many tasks in parallel for a single user. Large GPU clusters will always have the edge in efficiency.

← PreviousPage 2 of 34Next →