Agent harness choice strongly affects coding cost, not task success rate
Original titleAgent harnesses seem to make a lot of difference, at least for cost, even on open source coding benchmarks. Melissa’s research project di...
AISummary
Matei Zaharia says agent harnesses make a large difference in cost, even on open-source coding benchmarks, and Melissa Pan's research examines why.
Her quoted evaluation of seven models across Claude Code, Codex, and Pi found harness choice had little effect on task success but significantly affected cost. A simple harness can be competitive, and the native harness is not always the best.
Source: Matei Zaharia · x.comPublished · added here