raincole

8 ms·

31.
by raincole 8d ago
I'm confused. Isn't diffui using its own model?
32.
by raincole 8d ago
https://developers.openai.com/api/docs/pricing#image-tokens
33.
by raincole 8d ago
No way gpt-image-2 is as cheap as SDXL. (I dont concern about gpt-image-2's consumption either, just pointing out it's not a good comparison)
34.
by raincole 8d ago
Insert "Quit Having Fun" meme here.
35.
by raincole 8d ago
I feel it's more like a Nano Banana 2 update. It's much faster than gpt-image-2 but the quality isn't much different.
36.
by raincole 8d ago
Do you hate the designer changed the UI of your favorite app so you have to relearn it every three months? Congrats! Now you need to relearn it every time you open the app.
37.
by raincole 8d ago
Yes, I read the original statement. Buckmaster explicitly stated: > I have not seen OpenAI’s proof. I do not know what their model did, or how. I do not know whether our data was used. I am not accusing anyone of anything. People sayi
38.
by raincole 8d ago
This will be remembered as one of the biggest milestones in AI progress. The drama around it will at best be a footnote, just like hardly anyone caring about the drama around Poincare conjecture today.
39.
by raincole 8d ago
> People are saying OpenAI "stole" this from the work of an OpenAI user. If so, that's pretty fucked - how can we trust them? The said user (Tristan Buckmaster) didn't solve the millennium problem. He didn't real
40.
by raincole 8d ago
I totally agree. Powershell is a bit cryptic but it's really powerful especially on Windows. LLMs have smoothed out the learning curve.
41.
by raincole 8d ago
Off topic, but can anyone share their experience with DaVinci Resolve on Linux? It seems that they only officially support Rocky.
42.
by raincole 8d ago
When I still payed Dota, there was a wiki hosted on fandom. The experience was terrible. It wasn't always update-to-date with the latest patch, and on mobile there were so many ads rendering the whole thing unreadable. The catch is tha
43.
by raincole 9d ago
GPT4 was more expensive than Astra. And by the time it didn't even have cache. Tokens are so cheap and so commoditized today that people already forgot how things were in the earlier days of LLMs.
44.
by raincole 9d ago
I have a strong hunch this whole thing is just fiction written by LLM. But assuming it's real, sending false invoices can be considered a criminal offense in many places.
45.
by raincole 9d ago
I honestly don't get what the author is trying to say. > But pure mathematics is already in a conservatory, better known as the academy. Yes, exactly. So "is mathematics about to enter the conservatory" is a quite weird qu
46.
by raincole 9d ago
Playing go is art. Go was even considered one of the four main arts in ancient China. Along with calligraphy, instrumental music and brush painting[0]. It didn't stop AlphaGo from beating the best human go players, though. [0]: https:
47.
by raincole 10d ago
They release women from housekeeping... to workplaces.
48.
by raincole 10d ago
It's still very unclear to me how Anubis is supposed to work in the long term or even the medium term. The assumption is that scrappers, a.k.a. AI companies, a.k.a. those who hog all the RAM, are short of memory...?
49.
by raincole 10d ago
No, you explained why the selection bias doesn't matter, which I agree. The self-report data still is completely worthless in the first place though. People are really, really good at lying to themselves, let alone to an online poll. T
50.
by raincole 10d ago
It'd be more plausible if the current SOTA models weren't so good at generating SVG with all subjects instead of just pelican. We're witnessing the mission getting fucking accomplished [0]. [0]: https://xkcd.com&#x
51.
by raincole 10d ago
Wait, this is automatic? HN removes 'how'?
52.
by raincole 10d ago
"We invented a new term to express how much we dislike LLMs! Look at us, we certainly deserve your attention!."
53.
by raincole 10d ago
This is exactly how a company would act when they notice millions of people downloading and using their ram-hungry electron app. There is just no reason to not act this way.
54.
by raincole 10d ago
I've never encountered a case where I tell AI just uses plain JavaScript instead of React and it doesn't result in much cleaner code.
55.
by raincole 10d ago
(2025) Also in the author's latest post[0], they took this[1] self-reported preference poll seriously, which makes it very hard for me to take their articles seriously. [0] https://bcantrill.dtrace.org/2026/09&#x2F
56.
by raincole 10d ago
> obvious Who decide what is obvious or not?
57.
by raincole 11d ago
When I tried to post a duplicated link HN will just redirect me to the existing one, but somehow everyone else seems to be able to post the same link multiple times.
58.
by raincole 11d ago
AI writing just means "writing I don't like" now. Just like Nazi means whatever and whoever I politically disagree with. Words have lost their meaning.
59.
by raincole 12d ago
If you just "translate" an existing proof step by step to Lean, then of course you could mis-encode the intermediate statements too. But if you mis-encode the steps and still pass Lean check, it means you found a new proof! (Or yo
60.
by raincole 12d ago
The answer is we don't really know [0]: > In 2026, AIs designed to spot bugs in software were directed at Lean, and found several loopholes which were then fixed. Perhaps related to this effort, a purported disproof of the Collatz c

More ›