people · May 18, 2026
Cline Achieves 74.2% on Terminal-Bench 2.0 with Claude Opus 4.7
Share the canonical public link.
Cline ran Terminal-Bench 2.0 and revealed systematic differences for the same model across different runtimes on May 16, 2026. Cline gained 4.8 percentage points on opus-4.7 through runtime optimization compared to Claude Code at 69.4%. Cline's hill climbing experiment optimized harness prompts, tool definitions and context management from 47% to 57%. Cline extracted its internal runtime into an open-source SDK transforming from the underlying engine of a VS Code extension into reusable infrastructure.
Spend governor blocked model creation: provider_circuit_open (lane=battlemap, provider=together)