Claude Sonnet 5 (Non-reasoning, High): benchmark results

Provider: Anthropic. Released 2026-06-30. Access: API.

Unified ELO 1624 ± 1, rank #340 of 1761 rated models, from 17 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Artificial Analysis Intelligence Index33.49Intelligence Index85.8
AA Omniscience - Software Engineering (SWE)53.2Accuracy (%)82.1
AA Omniscience-0.65Score80.5
AA-Omniscience Accuracy33.8Accuracy (%)77.6
AA Omniscience - Humanities & Social Sciences31.6Accuracy (%)77.2
AA Omniscience - Science, Engineering & Mathematics39.28Accuracy (%)77.1
AA Omniscience - Business26.5Accuracy (%)76.8
AA Omniscience - Health31.48Accuracy (%)76.8
AA Omniscience - Law21.2Accuracy (%)72.1
AA Humanity's Last Exam19.05Accuracy (%)70.7
AA GDPval1288.23ELO69.2
AA GPQA Diamond80Accuracy (%)67.2

Interactive version: theaggregate.ai/model?slug=claude-sonnet-5-non-reasoning-high · How It Works · Data refreshed daily, snapshot 2026-09-05.