zurfer

6 ms·

1.
by zurfer 22d ago
Maybe try with OpenAi Luna. It's super cheap and not that stupid.
2.
by zurfer 22d ago
The same level of intelligence gets roughly 10x cheaper per year. So you might both be correct where a large part are commodity tasks but frontier is hard and valuable and not commodities.
3.
Tabular World Model Leaderboard(huggingface.co)
3 points by zurfer 24d ago|0 comments
4.
by zurfer 28d ago
Woah now also a 75perc discount on OpenRouter for flash 3.7. Is it really the same product (speed? and up time)? Why would Google do that?
5.
by zurfer 29d ago
But that's supply and demand, not technology. Right now a lot more people want their inference than they can supply. as supply catches up in the next 5-10 years, the underlying tech at scale is probably cheaper than GPUs per token pro
6.
by zurfer 1mo ago
maybe exactly because the frontier moves, introductory pricing makes sense as you want to free up compute for the newer models
7.
by zurfer 1mo ago
This is obviously cool and useful so kudos, but wow security researchers have to throw their hands up all the time. Now we open another attack surface where you can ask a remote agent to do things by default. There was a time when you call
8.
by zurfer 1mo ago
Zooming out you see that books, pen and paper are all technology. Like even school and language is technology. Given that AI exist maybe we can roll back some of the school/university and just let people do the work and learn while doi
9.
by zurfer 2mo ago
Well, we can't have the counterfactual world but could it be that without computers, Internet or AI, there might be no/less growth? I somehow doubt that AI will accelerate growth constantly as the world has inertia and I'm no
10.
by zurfer 2mo ago
If it's made for LLMs, the spec should be yaml not json. Way more token efficient
11.
by zurfer 2mo ago
It looks cool but how can I use it? Somehow i don't want to go hunting for access through the rabbit hole that is Microsofts Corporate blog.
12.
by zurfer 2mo ago
https://github.com/soofi-project/Soofi-Pretraining https://huggingface.co/Soofi-Project/Soofi-S-Base > "The final model will be released openly under a permissive license, without gated ac
13.
by zurfer 2mo ago
more interesting link: https://arxiv.org/html/2607.09424v2 and > Long-context serving efficiency. Soofi S combines frontier-level capability with the highest measured aggregate long-context decode TPS, and unlike f
14.
by zurfer 2mo ago
Somehow the blog post seems naive. Yes GLM 5.2 is good and cheaper per token, but margins are a result of supply and demand. Now demand for quality and quantity of tokens is increasing at least quadratic or cubic (more users * more tasks *
15.
by zurfer 3mo ago
> So he throws billions at a few top AI researchers, but they produce nothing of value. so he spends 1% of yearly revenue on AI talent to catch up? we can't judge if they have produced nothing of value, no? They don't owe the w
16.
by zurfer 3mo ago
ASML, SAP, ARM, Spotify, ... How do you define biggest? Can't be by market cap.
17.
by zurfer 3mo ago
This is called mechanistic interpretability. There is lots of fascinating insights already since you can do basically everything down to the neuron or weight level thousands of times. The human brain is many orders of magnitude harder to ma
18.
by zurfer 3mo ago
We outsourced it for 2.5k (extra) and it was still painful, took almost 2 months and worst of all wasted so much time and focus. The worst was sitting at the notary, and getting read out loud by her what we were about to sign (also paying f
19.
by zurfer 3mo ago
Like sqlite, duckdb is underappreciated as a production database. You can totally run it on servers or even "serverless" and do some heavy data transformations or with the right server size work with large scale datasets (up to a
20.
by zurfer 3mo ago
I heard the same argument from my doctor when I wanted a blood scan. But what's the intention? If you do a scan and then try to find everything that is wrong about you, you're 100% right, there will be false positives and unnecess
21.
by zurfer 3mo ago
Exactly, like SAP HANA stores everything in memory, you get great analytical and transactional performance but good luck financing that at scale
22.
by zurfer 3mo ago
It was a red herring.
23.
by zurfer 3mo ago
Is LTSE working the way you hoped?
24.
by zurfer 3mo ago
I do think it's more subtle. AI can replace very few jobs end to end with the same quality, maybe none. But AI can be put to work on high ROI problems. Now when the new marginal job is not obviously as high ROI as putting another 100k
25.
by zurfer 4mo ago
Don't fall for it. Its a trick question. Capitalism made everybody replaceable a long time ago. :P The more interesting question is: Has the ease of replacibility increased because of AI?
26.
by zurfer 4mo ago
Naive question but if it works best wouldn't companies that have a four day work week outperform theirs peers and because of that grow faster, and become more common? I see the opposite in most startups that have a 6 day work week to g
27.
by zurfer 4mo ago
Anthropic will probably IPO as well
28.
by zurfer 4mo ago
yeah 3t/s seems human. only that i never wrote code perfectly top to bottom.
29.
by zurfer 4mo ago
Kudos to the team. Please consider making the model available via API!
30.
by zurfer 4mo ago
In almost every wealthy country the birth rate has fallen below sustainable, meaning we're shrinking if life expectancy doesn't magically explode. Viruses btw never reflect on killing the host :P

More ›