CritPt v2 scores are generally higher than v1 and are not directly comparable. GPT-6 Astra (max) scores 83% on v2, up from 32% on v1. Both…

CritPt v2 scores are generally higher than v1 and are not directly comparable. GPT-6 Astra (max) scores 83% on v2, up from 32% on v1. Both changes contribute: in our testing, most of the increase comes from the revised challenges, with code-execution tool access adding a smaller gain. With leading models now above 80%, CritPt v2 leaves less headroom at the frontier than v1 did.