Recent submissions

Comment · 1 month ago
Comment · 1 month ago
Well it's not exactly saturated when OAI refused to use the harness explicitly provided by ARC-AGI. I'm not really familiar enough with the benchmark to declare whether it's a perfect measure for AGI, but I kind of doubt it is.
Comment · 1 month ago
Because they saw how much hype Glasswing was getting in April
Comment · 1 month ago
I'm always tired of seeing at the top of every new model release post on here. I say Simon should just keep it to Twitter.
Comment · 1 month ago
Don't forget the 3D demos. My favorite is in the house tour where the sink and stovetop(?) are obviously very misaligned from the counters
Comment · 1 month ago
72 to 74 on DeepSWE is AGI
Comment · 3 months ago
it seems the older models were capped at 10kusd for the runs though?
Comment · 4 months ago
Damn, that beam of light was a flashbang. I wouldn't call this tasteful UI design, but maybe I just need to go to sleep.
Comment · 4 months ago
Greentext is eh. Very formulaic, in fact very similar to the bottomless pit one, which I'd argue is better because of it's absurdity. I have to ask, did you mention the older GPT version to fable in the prompt?
Comment · 4 months ago
i got a 502
Comment · 4 months ago
Comment · 4 months ago
yeah man this sucks. i genuinely do not know how people find this stuff appealing
Comment · 4 months ago
I still don't know why people are saying this. I don't really code but from what I've heard on here the models haven't improved since Opus 4.5

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton