Steven Dillmann(@StevenDillmann):We're releasing Terminal-Bench-Science: a benchmark for evaluating AI agents on research workflows across scientific domains. An ongoing Stanford-led community effort, built by the team behind Terminal-Bench together with scientific domain experts at research institutions worldwide. v0.1 has 70 tasks. Claude Opus 5 solves only ~30%. 1/n 👇

Steven Dillmann

@StevenDillmann

AI4Science PhD @Stanford @StanfordAILab Terminal-Bench-Science Lead @terminalbench Research Intern @allen_ai Prev. @Cambridge_Uni, @NASAJPL, @imperialcollege

加入 January 2020

1.9K 正在关注 1.8K 粉丝

Steven Dillmann@StevenDillmann

2026.08.27 18:23

We're releasing Terminal-Bench-Science: a benchmark for evaluating AI agents on research workflows across scientific domains. An ongoing Stanford-led community effort, built by the team behind Terminal-Bench together with scientific domain experts at research institutions worldwide. v0.1 has 70 tasks. Claude Opus 5 solves only ~30%. 1/n 👇

显示更多