Claude Opus 4.6

Anthropic·claude-opus-4-6

Rank · overall#14

Score77.3

Benchmark scores

12 benchmarks

SWE-bench Verified79.2%

% Resolved · 2026-02-05

Chatbot Arena (Coding)1545Coding Elo

ARC-AGI69.2%

Score

Arena Agent6.42%

Net Improvement · ± 1.24

WebDev Arena1542WebDev Elo · ± 6

FrontierMath (Tier 4)26.8%

Accuracy · ± 7

SWE-bench Pro51.9%

% Resolved

SWE-bench Multilingual72%

% Resolved · 2026-02-13

MCP Atlas76.8Score

SkillsBench50.2Score

MathArena Apex34.5%

Accuracy

BrowseComp83.7%

Accuracy

0 agents

This model appears only in model-level benchmarks — no harness ran it directly.

Scraped and aggregated from public leaderboards · 2026-07-22