| Benchmark | Model | Rank | Results |
|---|---|---|---|
| agents-on-automationbench-private-v1-0-6 | GPT-6.1 Sol (max effort) (Zapier AutomationBench) | – | Success Rate: 36.1 |
| agents-on-terminal-bench-science-0-1 | GPT-6.1 Sol | – | Accuracy: 57.02 |
| code-generation-on-deepswe | GPT-6.1 Sol (high effort) (mini-swe-agent) | – | Pass@1: 0.7522 |
| computer-use-agents-on-osworld-2-0-0808-offline-82-tasks | GPT-6.1 Sol | – | Partial: 71.42 |
| medical-question-answering-on-healthbench-professional | GPT-6.1 Sol | – | Score: 64.2 |