Claude Opus 4.1

Benchmark Model Rank Results
agents-on-taubench-airlineClaude Opus 4.1–Accuracy: 56.0
agents-on-taubench-retailClaude Opus 4.1–Accuracy: 82.4
code-generation-on-humanevalClaude Opus 4.1–Pass@1: 95.7
code-generation-on-swe-bench-verifiedClaude Opus 4.1–Accuracy: 74.5
mathematical-reasoning-on-aime-2025Claude Opus 4.1–Accuracy: 78.0
multi-task-language-understanding-on-mmmluClaude Opus 4.1–Accuracy: 89.5
multimodal-reasoning-on-mmmuClaude Opus 4.1–Accuracy: 77.1
multiple-choice-question-answering-mcqa-on-gpqa-diamondClaude Opus 4.1–Accuracy: 80.9