Claude Sonnet 4.5 (Non-reasoning) Benchmark Results

Claude Sonnet 4.5 (Non-reasoning) — benchmark results

Claude Sonnet 4.5 evaluated with reasoning disabled. Provider: Anthropic. Released 2025-09-29. Access: API.

Unified ELO 1614 ± 9, rank #448 of 1776 rated models, from 306 benchmark results.

Benchmark	Score	Metric	Percentile
EuroEval Danish Common Sense Reasoning	90.41	Common Sense Reasoning Average Score (%)	100
EuroEval Dutch Common Sense Reasoning	93.05	Common Sense Reasoning Average Score (%)	100
EuroEval English Common Sense Reasoning	96.52	Common Sense Reasoning Average Score (%)	100
EuroEval French Common Sense Reasoning	87.88	Common Sense Reasoning Average Score (%)	100
EuroEval French NLU - Eltec	80.67	Named entity recognition Score (%)	100
EuroEval German Common Sense Reasoning	92.63	Common Sense Reasoning Average Score (%)	100
EuroEval Italian Common Sense Reasoning	91.54	Common Sense Reasoning Average Score (%)	100
EuroEval Spanish Common Sense Reasoning	94.95	Common Sense Reasoning Average Score (%)	100
EuroEval English Knowledge	97.47	Knowledge Average Score (%)	99.8
EuroEval Finnish Common Sense Reasoning	88.15	Common Sense Reasoning Average Score (%)	99.7
EuroEval Portuguese Common Sense Reasoning	98.65	Common Sense Reasoning Average Score (%)	99.7
EuroEval Swedish Common Sense Reasoning	90.52	Common Sense Reasoning Average Score (%)	99.7