mattnewton

5 ms·

31.
by mattnewton 4mo ago
In the US, I think we are being intentionally DDOS-ed. This strategy was laid out by Steve Bannon in the old frontline PBS interview where he called the media “the opposition party.” “They’re dumb and they’re lazy, they can only focus on on
32.
by mattnewton 4mo ago
Yeah people are overdoing it, and not everyone is being motivated to do something by feeling mad, but maybe instead of throwing up your hands that you can’t care about soybean tariffs, you could try to educate yourself on tariffs and choose
33.
by mattnewton 4mo ago
_in a polling place_ no less
34.
by mattnewton 4mo ago
Definitely encourage you to test the models. We tried to optimize for realistic focus and not over-sharpening, which leads to a "hyper" AI-look. It's hard to benchmark because people generally prefer sharp, saturated orangis
35.
by mattnewton 4mo ago
Krea 2 Large (on the website and api) was trained with the FLUX 2 VAE, if you want to test it out and push realism. After working with both I think the flux VAE has a slight edge in learning realistic textures but it's smaller than you
36.
by mattnewton 4mo ago
You can find some links and details in the GitHub readme for finetuning / LoRA support. Ostiris, musubi tuner, fal and hugging face diffusers are all day-0 supported :) https://github.com/krea-ai/krea-2 We recomm
37.
Krea 2: SOTA open-weights 12B image model(krea.ai)
415 points by mattnewton 4mo ago|47 comments
38.
by mattnewton 4mo ago
Hi HN, we're releasing weights for our latest text to image model and publishing this writeup on how we trained it in quite a bit of depth. I hope there is something in the report for everyone, we included a fair bit on the actual trai
39.
by mattnewton 4mo ago
The goal is to encroach on privacy, and those interests are using the children able to do it. There is no “solution” to the children using the internet anonymously problem that will satisfy them. We have to keep fighting for privacy.
40.
by mattnewton 4mo ago
I do not need to have the real secret of immortality to say to the emperor that swallowing mercury is bad. The burden of proof is on those who put forward a solution.
41.
by mattnewton 4mo ago
In my view, it’s very simple. There are places like schools, or parents buying phone plans, to identify children. Will some children get access outside of that? Sure. But 100% enforcement isn’t possible even if you thought it was worth dest
42.
by mattnewton 4mo ago
Because banning smartphones in schools doesn’t affect adults not in those schools, whereas age verification does? How you implement these protections matter.
43.
by mattnewton 4mo ago
They have a few in models that powered the app, including one that applied edits. Recently they also started fine tuning Kimi models under the “composer” brand, and Composer 2.5 is a very cost effective coding model. But I suspect that the
44.
by mattnewton 4mo ago
I recommend watching the video, he makes an IMO excellent case that #1 without a lawyer really is a bad idea. You can still help solve the case in way that protects you, the stakes here are often incredibly high.
45.
by mattnewton 4mo ago
Congrats to the Cursor team! When I first saw the company built on top of vscode in such a crowded field way back at the end of 2022, I thought "forget having a moat, they are renting their castle from the invaders!" - I couldn&#x
46.
by mattnewton 4mo ago
they have had one for a while now. https://cursor.com/cli
47.
by mattnewton 4mo ago
I think it is less about competing for athletes and more about competing for national attention (in the form of sports viewership that turns into money and school programs).
48.
by mattnewton 4mo ago
I don’t disagree with your conclusion that this is likely ai rewritten, but I do find it strange that you say “normal people don’t write like this” when it is mimicking how people write, and using patterns I have seen people write. I think
49.
by mattnewton 4mo ago
this is almost certainly too recent to have been used for training data, no? Unless they optimistically included most repos somehow?
50.
by mattnewton 4mo ago
No insider info, but just wanted to mention that pricing signals things too. If Mythos is only servable at $X*Y dollars and isn’t Y times better than $X of compute at another provider, it’s quite possible that affects the IPO price negative
51.
by mattnewton 5mo ago
I wouldn't be surprised if each of the frontier American labs and individually has compute access similar to the entire EU. Chinese firms are a more interesting comparison since there are a fair amount of great models there, and it&#x2
52.
by mattnewton 5mo ago
My theory with no insider information: it’s a little of all of the above, but mostly money. To some extent, you can dig yourself out of a data hole with RL and a lot of compute. And you can buy a lot of compute and some data with a lot of m
53.
by mattnewton 5mo ago
Litmus tests about personal beliefs are not really how religious organizations function for most people in my experience. It’s about whether you want to associated with a tribe or movement, then the beliefs come with that package.
54.
by mattnewton 5mo ago
But they also have 1,000,000 of expenses, at least some of which is probably deductible from income.
55.
by mattnewton 5mo ago
They are betting they can sell the bag before the music stops.
56.
by mattnewton 6mo ago
> or instance, maybe you can't afford to take on more customers right now, Anthropic. Maybe if you are severely undermining the customer relationships you already have, you should just admit you can't sell any more 20x plans ri
57.
by mattnewton 6mo ago
Idk about morality, but it’s certainly a way to stop dystopian mass surveillance nightmares if everyone capable of building one refuses. So if you live in the US and don’t want one government agency in the US to have this power (that is amb
58.
by mattnewton 6mo ago
My best guess is that for certain goods, people make purchasing decisions around major life events like getting married or becoming parents. If they have already crossed those thresholds those purchase patterns may be harder to unseat and r
59.
by mattnewton 6mo ago
people are trying, especially for inference. For training, it’s just too high risk to tank your training I think. TPUs are at least dogfooded by Google deepmind, no team AFAIK has gotten the AMD stack to train well.
60.
by mattnewton 6mo ago
https://www.theguardian.com/lifeandstyle/2025/jul/07/no-safe... > Now thought to be the second most common cause of stroke in women under 40, it can also lead to difficulty swallowing, incontinence, s

More ›