Claude 3.7 Sonnet and Claude Code

Benchmark Model Rank Results
agents-on-taubench-airlineClaude 3.7 Sonnet–Accuracy: 58.4
agents-on-taubench-retailClaude 3.7 Sonnet–Accuracy: 81.2
code-generation-on-swe-bench-verifiedClaude 3.7 Sonnet–Accuracy: 62.3
mathematical-reasoning-on-aime24Claude 3.7 Sonnet–Acc: 23.3
mathematical-reasoning-on-math-500Claude 3.7 Sonnet–Accuracy: 82.2
multi-task-language-understanding-on-mmmluClaude 3.7 Sonnet–Accuracy: 83.2
multimodal-reasoning-on-mmmuClaude 3.7 Sonnet–Accuracy: 71.8
multiple-choice-question-answering-mcqa-on-gpqa-diamondClaude 3.7 Sonnet–Accuracy: 68.0