Thread

1. BoumTAC · 1790013700 · focus · HN ↗

   Vals AI just affirm that Grok 4.7 is worse than Grok 4.6 (It ranks #24 on the Vals Index at 54.2%, down 5.0 points from Grok 4.6 (#14, 59.2%))

   <a href="https:&#x2F;&#x2F;x.com&#x2F;ValsAI&#x2F;status&#x2F;2102086608476590432" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;ValsAI&#x2F;status&#x2F;2102086608476590432

   1. nostrebored · 1790013882 · focus · HN ↗

      Is this an ad for Vals AI? Looking at their website, the rankings don&#x27;t mesh with my observed utility for almost any model outside of fable and astra being good-ish.

      1. BoumTAC · 1790014043 · focus · HN ↗

         Absolutely not. I know Elon retweet them a lot when Grok is good. This is how I discover the company.

         I like to follow them and look for benchmark for each LLM release.

         1. nostrebored · 1790019102 · focus · HN ↗

            Ah gotcha, not on twitter so just hadn&#x27;t seen them before!

Unofficial Hacker News client; not affiliated with Y Combinator.