Claude Sonnet 5 (Max) — benchmark results

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

Unified ELO 1985 ± 37, rank #27 of 1744 rated models, from 7 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Epoch AI - Scicode53.59Score88
Epoch AI - Critpt16.86Score83.1
Epoch AI - Cursorbench61.2Score65.4
FrontierMath - Tiers 1-3 (v2)65.61Accuracy (%, 285 private v2 problems)62.9
CursorBench 3.161.5Score (%)61.3
FrontierMath - Tier 4 (v2)29.27Accuracy (%, 41 private v2 problems)59.7
AutomationBench13.5Pass Rate (%)0

Interactive version: aibenchmarks.dev/model?slug=claude-sonnet-5-max · How the rankings work · Data refreshed daily, snapshot 2026-07-10.