Artificial Analysis Launches Intelligence Index v5

Artificial Analysis unveils Intelligence Index v5, its biggest upgrade since January, featuring 60% based on private test sets. The release includes FrontierCode 2.0 and Terminal-Bench-Science benchmarks. Claude Opus 5.5 leads the rankings, while open weights models widen their gap to 12 points behind the leader. Costs vary by 13x among top models, ranging from $0.96 to $13.04 per task.