GPT-5.5 – No ARC-AGI-3 scores

9 points | by AG25 13 hours ago ago

2 comments

  • LouisLau 6 hours ago ago

    ARC-AGI-3 is interesting, but it’s not necessarily aligned with what models are being optimized for right now (reasoning + usefulness in real tasks).

  • casey2 11 hours ago ago

    ARC-AGI-3 scoring is really weird, in some views it's already saturated in others it's near 0. But I assume, since the entire benchmark IMO is a PR tool for OpenAI they will publish it eventually.