Artificial Analysis Intelligence Index v5
Summary
Artificial Analysis announces Intelligence Index v5 with two new benchmarks and three upgrades, now largely evaluated on private datasets, plus the new Intelligence Progress Index v1. The piece details benchmark updates (FrontierCode 2.0, Terminal-Bench-Science, HLE-Diamond, AA-Omniscience v1.1, CritPt v2) and compares leading models like Claude Opus 5.5 and GPT-6 Astra, including cost per task and token usage. It also links to methodology and encourages exploring the full results.