AIFIRST Terminal-Bench-Science: Evaluating AI agents on scientific research workflows
The brief
Terminal-Bench-Science is a new benchmark for evaluating AI agents on scientific research workflows executed entirely in a terminal environment.
Key points
- Tasks include literature search, data analysis, hypothesis generation, and experimental design, assessed against expert researcher baselines.
- The benchmark is designed to measure autonomous research capability rather than isolated question-answering performance.
Sources
- HNterminal-bench-science.ai