Home Model series Claude Sonnet Claude Sonnet · GDPval-AA · v2 · Release history Claude Sonnet · GDPval-AA · v2 · Release history View scores, configurations, and disclosures for GDPval-AA · v2 across 2 Claude Sonnet releases, ordered by release date.
Differences are calculated only for unique, active, exact values between direct predecessors when the unit, metric, and explicitly recorded configuration and conditions match.
2026-02-17 · 1 score record for this release, including 1 exact value (may include historical records)
Claude Sonnet 4.6 · GDPval-AA scores and configurations Metric and score Configuration and evaluation conditions Sources Elo rating 1633 Elo Reported 1633 · Unit Elo Published Unconfirmed
Evaluation settings not fully stated Settings Only settings explicitly listed in each result's conditions are known. Unknowns Effort, sampling and tool settings are unspecified unless stated in result.conditions. Tools Tools allowed Reasoning budget not recorded Framework not recorded Protocol 220 GDPval tasks across 44 occupations and 9 industries; blind pairwise comparisons. Conditions Models receive shell access and web browsing in an agentic loop. Effort and sampling details are not stated in the source excerpt. View this score in rankings
2026-06-30 · 1 score record for this release, including 1 exact value (may include historical records)
Records for the direct predecessor do not provide matching conditions for a valid difference.
Claude Sonnet 5 · GDPval-AA scores and configurations Metric and score Configuration and evaluation conditions Sources Elo rating 1618 Elo Reported 1618 Elo · Unit Elo Published Unconfirmed
Evaluation effort not specified Settings Only the setup explicitly named in the result's conditions is known. Unknowns Effort and sampling may be unknown; see each result.conditions. Tools Tools allowed Reasoning budget not recorded Framework not recorded Protocol GDPval-AA v2, 220 tasks across 44 occupations and 9 industries; blind pairwise comparisons. Conditions Shell and web browsing in an agentic loop; evaluation independently run by Artificial Analysis. View this score in rankings