| agents-on-terminal-bench-science-0-1 | Claude Haiku 5.5 (max effort) (Claude Code) | – | Accuracy: 20.6 |
| chart-question-answering-on-chartography | Claude Haiku 5.5 (with tools, max effort) | – | Pass@1: 0.862 |
| chart-question-answering-on-chartography-no-tools | Claude Haiku 5.5 (max effort) | – | Pass@1: 0.464 |
| code-generation-on-frontiercode-1-1-main | Claude Haiku 5.5 (max effort) (Claude Code) | – | Score: 46.4 |
| code-generation-on-programbench-golden-166-no-time-limit | Claude Haiku 5.5 (max effort) (mini-swe-agent) | – | Average Test Pass Rate: 82.0 |
| code-generation-on-swe-bench-multilingual | Claude Haiku 5.5 (max effort) | – | Accuracy: 83.7 |
| code-generation-on-swe-bench-multimodal | Claude Haiku 5.5 (max effort) | – | Accuracy: 30.7 |
| code-generation-on-swe-bench-pro | Claude Haiku 5.5 (max effort) | – | Resolved: 64.8 |
| code-generation-on-terminal-bench-4-0 | Claude Haiku 5.5 (max effort) (Claude Code) | – | Accuracy: 39.2 |
| medical-question-answering-on-healthbench-opus-4-8-grader-length-adjusted | Claude Haiku 5.5 (max effort) | – | Score: 61.6 |
| medical-question-answering-on-healthbench-professional | Claude Haiku 5.5 (max effort) | – | Score: 64.8 |
| multi-task-language-understanding-on-global-mmlu | Claude Haiku 5.5 (max effort) | – | Accuracy: 87.8 |
| multi-task-language-understanding-on-milu | Claude Haiku 5.5 (max effort) | – | Accuracy: 87.6 |
| question-answering-on-officeqa-pro-preselected-text | Claude Haiku 5.5 (max effort) | – | Accuracy: 60.3 |