← All benchmarks

Terminal-Bench 2.1 Extended (Mercor held-out)

Code

Mercor's held-out Extended twin of Terminal-Bench 2.1: 99 Mercor long-horizon command-line agent tasks, scored by task accuracy. The public twin is Terminal-Bench 2.1 (terminal-bench-2-1). Mercor does not publish the held-out tasks or say how they were drawn or written, so this is its own benchmark, not a component of the public one.

Top models (higher is better)