Hacker Newsnew | past | comments | ask | show | jobs | submit | codemk8's commentslogin

Same feeling. It’s great when GPUs are available. Also helps if you can use non data center cards in your workflow.

This post exactly ;-)


Me.


If I had an oracle, why would I need an LLM?


You mean deepseek-v4-flash, right? Same here. I use it for my Hermes agent. It's so cheap that I sometimes feel "guilty". I even put more money than I needed just make sure they do not go out of business.


Yes, I do mean deepseek-v4-flash.


Runpod with its python APIs.


From my recent experience, ROCm and hipcc allow you to port cuda programs fairly easily to AMD arch. The compiling is so much faster than nvcc too.


This reminds me of the user interface for pied piper’s compression algorithm

https://images.app.goo.gl/G1K7J8gZ135x98mz7


Expecting the Netflix show of this book though.


Mild Spoilers:

I think that'll be a pretty hard book to adapt. The story goes off the rails in the later books. Especially the 3rd, it's just a random tour of interesting scifi ideas.

I'm also wondering if the de-emphasis of individual will make it harder to make a show. There is no through line of character.

While the book has some great ideas, I personally wasn't a huge fan of how they were strung together. First book was great detective noir book. Second book had the great concept of the wallfacers and one amazing badass. And then you can just see disaster coming with the change of the guard, the story is just as bad as the end of the rama series from there.

Or to say it another way, if it makes it to the screen it will be heavily altered imho, but that's par for the course, and I've come to enjoy other people's interpretations on things I've already read when they bring new stuff.


Lol I posted the same link 2 mins after u


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: