alt Hacker NewsWow, 30% on ARC-AGI-3 for $20k total. Huge jump from GPT-5.6's 7.8% at $20k per task. I continue to believe ARC-AGI measures something different and important compared to other benchmarks.
Replies
dominotw • yesterday at 5:45 PM
> I continue to believe ARC-AGI measures something different
why is that? its now being benchmaxxed too
seeing a jump this big is not a great sign for the continuing value of a benchmark