Parallel beats serial at equal budget: best-of-N@T far outperforms best-of-1@N*T

ShangyinT · x · 2026-09-22

In a debate on multi-agent baselines, generatorman argues that the fair baseline for team-of-N with total token budget T is not best-of-N@T but best-of-1@NT — the same budget spent serially on one agent.

ShangyinT reports that for reasonably large NT, best-of-1@NT is much worse than best-of-N@T, i.e. parallel sampling/agents beat stacking the budget serially. DimitrisPapail confirms this is measurably true across all his non-science projects.

Related event: Multi-Agent vs Serial Baseline Debate: Team Win Said to Need 10-100x Compute(8 posts)→

Original post →

More from coding & agent

coding & agent channel →