Haaargio

3 ms·

1.
by Haaargio 9mo ago
communication before the internet was very slow. Hype or getting viral is not necessarily science so its not clear when and how and why one paper suddenly becomes very known. We know what sugar and others do, people are probably ignorant or
2.
by Haaargio 9mo ago
Or why not just buy a blackwell rack? Runs everything today with bleeding edge performance. Overall whats the difference between 8k or 30k? /s
3.
by Haaargio 9mo ago
We are in a free market with china still playing the open source game. The market is not ready for building this due to costs etc. not because the big companies block them or anything. And nvidia is not selling subscriptions at all.
4.
by Haaargio 9mo ago
Its not dying slowly right now at all. Compute is a massive driver for everything ML. From number of experiments you can run in paralle, to how much RL you can try out, how long stuff is running etc. ML is pushing scaling on dimensions we h
5.
by Haaargio 9mo ago
Get yourself an IP with Port 443 free and just use that. SMTP is blocked by a lot of firewalls by default. All cloud providers do that and you need to request opening them up.
6.
by Haaargio 9mo ago
A very nit picky comment. In avg, the normal way it creates the liability over time and i would argue that in a colloquial its absolutly fine and doesn't change the message at all.
7.
by Haaargio 9mo ago
I would do the same thing if I would justifing paying 200$ per Month for my hobby. But even with that, you will run into throttling / API / Resource limits. But AI Agents need time. They need a little bit of reading the sourcecode