List your company

Terminal Task Execution on TerminalBench 2.0 (test)

35.2Test AccuracyCRO
31.0432.1233.234.28May 11, 2026

Evaluation results

MethodLinks
CROWall (minutes)=732026.0535.2
Baseline2026.0531.2
GEPAWall (minutes)=1572026.0531.2
MetaHarnessWall (minutes)=1732026.0531.2