Recent submissions

Comment · 2 months ago
haiku 4.5 top 7, that benchmark is absolute crap
Comment · 2 months ago
How I tackle complex problems with benchmark driven development.
Comment · 3 months ago
Yes I agree, but I actually did a lot more runs, with different prompts, different times ect... And each time /goal had a small or insignificant impact
Comment · 3 months ago
well thank you so much for this
Comment · 3 months ago
A deepdive on the /goal effect on a problem literally made for this.
Comment · 3 months ago
It seems that it is very strong at Frontend, I wonder how good are its multimodal capabilities. This is very important for me and not well represented in benchmarks in my opinion.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton