AgentBench

AgentBench copied to clipboard

**Describe the bug** cert expired **To Reproduce** go to llmbench.ai **Screenshots or Terminal Copy&Paste**

jlsajfj

dannyslowpark

有docker的替代方案吗？ podman k8s 没有sudo 权限

YerongLi

[Assistance] OS task retrun infos

1

**Describe the bug** 我看我们的OS环境目前是以用docker的方式把环境进行打包，之后利用Python的Docker.py和os环境交互。 想问一下，我们的OS环境，有相关的操作说明吗？ 就类似RL的环境 每一步都有哪些可执行操作、环境会有的反馈以及每一次的reward，我想单独把OS交互的代码拿出来，之后适配到我们自己的任务里……

xiaxiaxiatengxi

Hello, Do you make model trajectories (and their interactions with the system available)? I couldn't find them. Thank you!

felipemaiapolo

[Bug/Assistance] 在测试ltp的时候遇到了INTERACT_FAILED的问题

1

**Describe the bug** 在测试ltp环境的时候遇到了INTERACT_FAILED的问题，进度条一直显示为0 **Screenshots or Terminal Copy&Paste** **Desktop (please complete the following information):** - OS: MacOS - Python: 3.9

adoptedirelia

Operation system dataset has wrong start script. 1. dev.json "description": "what is the output if i execute ~/test?" the code should not be run as init script but as start...

Shuwen27