Recent submissions

Comment · 22 days ago
I really thought UTF-8 was already the default.
Comment · 1 month ago
That benchmark could really be a good AGI test. Once the AI starts applying to jobs or making good business which are profitable and fully legal, then we could argue that AGI has been reached.
Comment · 2 months ago
Sure, but you can do that at home too.
Comment · 2 months ago
Ask the AI to create a detailed spec according to a few simple requirements. Review the spec yourself and correct what you want changed. Then ask the AI to implement the spec. Each time you request something new, ask the AI to update the spec as well.
Comment · 5 months ago
I just completed the course and a big part of it was in fact to warn not to use it to send private data.
Comment · 5 months ago
No, it works using Malta eID identification system which needs you to be a citizen or a resident.
Comment · 5 months ago
Why are tiles small BTW? Could we use tiles as big as normal solar panels?
Comment · 5 months ago
Comment · 5 months ago
Like the Delamain AI in Cyberpunk. You would need to allow anonymous payments with cryptocurrencies for that, but it's coming for sure.
Comment · 6 months ago
For open-weights models, censorship removal is now a "solved" problem. If you wait a few days after a new model release, someone will have made a heretic ( https://github.com/p-e-w/heretic ) version with the censorship removed, so in a way the only use for censorship now is to avoid lawsuits, not reduce improper usage.
Comment · 6 months ago
Please mention and support llama.cpp directly instead of ollama.
Comment · 6 months ago
Why not try to find a successor instead of archiving the repo and forbidding the use of the name? I'm sure with a 3.8k stars repo you'll find competent people willing to continue the work.
Comment · 6 months ago

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton