Code Evaluation Shows Harness Impacts Cost

HankYeomans · x · 2026-07-09

The post references an evaluation based on real-world coding tasks: for the same model, choosing a different harness can result in a cost difference of about 2x. It also notes that Pi outperforms Claude Code in certain tests, and when paired with GLM 5.2, it delivers better results at a lower cost.

Related event: Optimizing Agent Harnesses Cuts Costs More Than Upgrading Models(6 posts)→

Original post →

More from coding & agent

coding & agent channel →