Claude 3.7 Sonnet

anthropic · Ranked across 6 benchmarks · best rank #9

Benchmark scores

BenchmarkCategoryRankScoreCaptured
Aider Polyglot code #9 64.9% 2025-02-24
METR Time Horizon agents #13 1h 2025-02-24
SWE-bench Verified agents #18 66.4% 2025-05-14
AA Coding Index code #33 27.6% 2026-06-15
Chatbot Arena chat #46 1314 2026-06-10
OpenRouter · Weekly Usage usage #57 #568 2026-06-09