Benchmarks
Terminal-Bench
A benchmark of hard, realistic tasks that AI agents must complete in a Linux command line, each in its own container and checked by tests. Now in version 4.1.
Tag
Command Line on The Latent Daily: 1 other resource. News, explainers and guides on Command Line, each linked to its sources.
A benchmark of hard, realistic tasks that AI agents must complete in a Linux command line, each in its own container and checked by tests. Now in version 4.1.