GPT-6 Astra Sets ARC-AGI Records, Scoring 99.9% on ARC-AGI-3

GPT-6 Astra set new records across all three ARC-AGI benchmarks, scoring 62.7% in the standard harness and 99.9% on ARC-AGI-3, with costs far below what Opus 5 spent to reach just 30% a month earlier, according to the ARC Prize team.

2026-09-04 ~ 2026-09-04 · 3 related posts