正在加载模式…
⌨️
Terminal-Bench(TB)
Hard command-line tasks in isolated environments, each with a human-written solution and tests that decide whether the agent actually finished.
复杂度: high评估与监控
30秒速览
- 是什么
- Poses hard command line tasks in isolated containers, each carrying a human-written solution and tests that decide afterwards whether the work was actually finished.
- 何时使用
- You are evaluating agents that do infrastructure, build or debugging work in a shell, and you need correctness decided rather than judged.
- 注意
- The score names a model and a harness together: the same weights under a weaker scaffold give up exactly where a stronger one reads the error and recovers.
向AI专家咨询此模式
打开助手并预填您的问题,发送前可先确认。
Terminal-Bench: 概览
Hard command-line tasks in isolated environments, each with a human-written solution and tests that decide whether the agent actually finished.
- One container per task, so runs cannot contaminate each other
- Every task carries a human-written reference solution
- Verified by tests rather than by inspecting the transcript
- Drawn from real infrastructure, data and debugging work
- Deliberately hard: frontier systems remain well short of solving it
- Harness-agnostic, so the agent and the scaffold are measured together
获取智能体评估实战指南
25 种智能体评估方法浓缩成一份指南：每个基准测量什么、公开分数何时具有误导性、如何用自己的失败构建评估。链接随确认邮件送达，并附每周的 The Agent Architect。
每周一封邮件,一键退订。您的地址仅用于发送简报。
又称： CLI agent benchmark, Command line agent evaluation
参考资料
该模式所依据的论文、规范和代码仓库。
由本目录背后的工程师执行
看看你的评测漏掉了什么
评测智能体比做出来更难，多数测试套件一路全绿，线上却在悄悄漂移。我们从头到尾检查你的评测体系：现在测到了什么、还有什么看不见，以及当前套件会放过哪些回归。
€750（原价 €1,500），一周交付，书面报告加讲解通话，9月30日前