
Terminal-Bench
A benchmark for terminal agents
AI researchers, machine learning engineers, and data scientists interested in novel LLM evaluation methods or domain‑specific language understanding
**SkateBench ranks 44 large language models on their ability to name skateboarding tricks correctly out of 210 tests.** Top performers include o3-pro, Gemini-3-Flash-High, and GPT-5.1-Default, with success rates visualized from 0% to 100%.