Comment

Comment by jstummbillig

Is there a non-obvious reason that models keep getting compared to GPT-3.5 instead of 4?

Discussion

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton