| chart-question-answering-on-chartography | Claude Sonnet 5.5 (with tools, max effort) | #1 | Pass@1: 0.902 |
| code-generation-on-terminal-bench-4-0 | Claude Sonnet 5.5 (Claude Code) | #1 | Accuracy: 70.6 |
| medical-question-answering-on-healthbench-opus-4-8-grader-length-adjusted | Claude Sonnet 5.5 (max effort) | #1 | Score: 65.4 |
| medical-question-answering-on-healthbench-professional | Claude Sonnet 5.5 | #1 | Score: 69.2 |
| agents-on-automationbench-private-v1-0-6 | Claude Sonnet 5.5 (max effort, with fallbacks) (Zapier AutomationBench) | #2 | Success Rate: 44.7 |
| agents-on-terminal-bench-science-0-1 | Claude Sonnet 5.5 (Claude Code) | #2 | Accuracy: 59.9 |
| code-generation-on-programbench-golden-166-no-time-limit | Claude Sonnet 5.5 (max effort) (mini-swe-agent) | #2 | Average Test Pass Rate: 79.7 |
| code-generation-on-swe-bench-multimodal | Claude Sonnet 5.5 (max effort) | #2 | Accuracy: 54.3 |
| code-generation-on-swe-bench-pro | Claude Sonnet 5.5 | #2 | Resolved: 81.3 |
| computer-use-agents-on-osworld-2-1-108-tasks | Claude Sonnet 5.5 (max effort) | #2 | Partial: 80.1Binary: 43.5 |
| multi-task-language-understanding-on-global-mmlu | Claude Sonnet 5.5 (max effort) | #2 | Accuracy: 92.1 |
| multi-task-language-understanding-on-milu | Claude Sonnet 5.5 (max effort) | #2 | Accuracy: 91.6 |
| question-answering-on-officeqa-pro-preselected-text | Claude Sonnet 5.5 (max effort) | #2 | Accuracy: 65.6 |
| code-generation-on-swe-bench-multilingual | Claude Sonnet 5.5 | #3 | Accuracy: 90.3 |
| agents-on-toolathlon-verified-108-tasks | Claude Sonnet 5.5 | #4 | Pass@1: 0.778 |
| code-generation-on-deepswe | Claude Sonnet 5.5 (max effort) | #9 | Pass@1: 0.71 |