| agents-on-automationbench-private-v1-0-6 | Claude Fable 5.1 (max effort, with fallbacks) (Zapier AutomationBench) | – | Success Rate: 31.4 |
| agents-on-terminal-bench-science-0-1 | Claude Fable 5.1 (max) (Claude Code) | – | Accuracy: 52.6 |
| code-generation-on-deepswe | Claude Fable 5.1 (max effort) | – | Pass@1: 0.674 |
| code-generation-on-programbench-golden-166-no-time-limit | Claude Fable 5.1 (max effort) (mini-swe-agent) | – | Average Test Pass Rate: 87.6 |
| code-generation-on-swe-bench-multilingual | Claude Fable 5.1 (max effort) | – | Accuracy: 89.1 |
| code-generation-on-swe-bench-multimodal | Claude Fable 5.1 (max effort) | – | Accuracy: 54.7 |
| code-generation-on-swe-bench-pro | Claude Fable 5.1 (max effort) | – | Resolved: 81.2 |
| code-generation-on-terminal-bench-4-0 | Claude Mythos 5.1 (max) (Claude Code --bare) | – | Accuracy: 60.9 |
| medical-question-answering-on-healthbench-opus-4-8-grader-length-adjusted | Claude Fable 5.1 (max effort, with fallbacks) | – | Score: 60.0 |
| multi-task-language-understanding-on-global-mmlu | Claude Fable 5.1 (max effort) | – | Accuracy: 94.0 |
| multi-task-language-understanding-on-milu | Claude Fable 5.1 (max effort) | – | Accuracy: 93.0 |
| question-answering-on-officeqa-pro-preselected-text | Claude Fable 5.1 (max effort, with fallbacks) | – | Accuracy: 69.0 |