Terminal Bench 2.0

Terminal Bench 2.0 is a benchmark framework for evaluating AI models' programming and operational capabilities in terminal environments, representing the second major version iteration. The benchmark designs tasks involving command-line operations, script writing, and system interaction to measure models' code generation, debugging, and execution abilities in real terminal scenarios, providing a standardized basis for cross-system performance comparison.