Claude Sonnet 3.7-64K Extended Thinking Benchmark Details
Claude Sonnet 3.7-64K Extended Thinking currently shows benchmark results led by GPQA Diamond (49 / 177, score 84.80), MATH-500 (19 / 44, score 96.20), AIME 2024 (27 / 62, score 80).
Benchmark Results
Claude Sonnet 3.7-64K Extended Thinking