Skip to content

Coding and terminal agents

Terminal-Bench 4.0

Current continuous Terminal-Bench release with task and resource revisions; catalog snapshot lists 66 tasks.

About Terminal-Bench 4.0

What it measures

Verifier-defined success under the versioned runtime budget.

Scope & limitations

The release sets an eight-hour agent timeout. Use the exact v4.0.0 task manifest; a shorter-budget run needs its own label. 3.0 and 2.1 remain separate.

Evaluation environment

Harbor terminal agent; some tasks need GPUs or multiple containers.

Source

Original benchmark

Dataset & task catalog ↗