Thread

1. leobuskin · 1789919390 · focus · HN ↗

   Astra had an explicit medium/xhigh levels, Fable - just Fable. What reasoning level was used? Why not multiple were tested?

   I’ve scrolled the article, but haven’t noticed any remarks about Fable’s levels.

   1. monk_grilla · 1789952170 · focus · HN ↗

      I was looking for this comment. Why were only the Codex models exercised at each effort level?

      1. steve_taylor · 1790082055 · focus · HN ↗

         We're probably seeing only the cherry-picked results.

Unofficial Hacker News client; not affiliated with Y Combinator.