Skip to content

Anthropic

Claude Opus 4.1

A small upgrade to Opus 4.

Announced
Access
Public

Benchmark scores

TestScoreBest on the trackerHow it was measured
Humanity's Last Examexpert questions, no tools12%GPT-6 Astra 55%Independentleaderboard entry "claude-opus-4-1-20250805-thinking" (claude-opus-4-1-20250805_unknown)
FrontierMath T1–3hard maths problems13%Claude Fable 5.2 95%Independent32K thinking budget (claude-opus-4-1-20250805_32K)Run 11 Jun 2026
GPQA DiamondPhD-level science questionsPhD experts: 70%Saturated77%Gemini 4 Argon 96%Independent16K thinking budget (claude-opus-4-1-20250805_16K)Run 5 Aug 2025
FrontierMath T4research-level maths problemsSaturated2.4%GPT-6.1 Sol 100%Independent32K thinking budget (claude-opus-4-1-20250805_32K)Run 11 Jun 2026
OTIS Mock AIMEcompetition mathsSaturated69%GPT-5.5 100%Independent27K thinking budget (claude-opus-4-1-20250805_27K)Run 5 Aug 2025
SWE-bench Verifiedfixing real software bugsKnown flaws73%Claude Opus 4.7 83%Independentdefault settings (claude-opus-4-1-20250805)Run 11 Feb 2026
METR time horizontask length an AI agent finishes half the timeNo longer updated1.7 hClaude Mythos Preview 17 hIndependentMETR Horizon v1.1 run 'claude_4_1_opus_inspect'; 95% CI 59.3–159.5 min

Compare every model on the same tests →

Sources