CLAIM-2026-10-14-02

On CRBugBench the best agent, claude-opus-5 with copilot-cli, catches 71.9% of bugs within its top five comments, up from 47.7% for opus-4.6 over five months.

Why it matters

A twenty-four-point jump in five months is the number that will carry a slide. It is also the number that makes a procurement decision look safe for longer than it is.

Sources

• arxiv2026when