研究：AI智能体团队耗费巨量token却几乎无质量提升
原标题：AI agent teams waste massive tokens for barely measurable quality gains, research finds
AI 摘要
评测公司 Vals AI 在 Vibe Code Bench 上测试了 GPT-6 Sol 和 Claude Opus 5.5 的单体与多智能体团队表现，发现团队成本是单体的 1.8 至 5.1 倍，但四次对比中仅一次有统计显著提升。Anthropic 自身测试也显示，从 10 个扩展到 100 个智能体后分数仅微增。OpenAI 研究员 Noam Brown 和开发者 Eric Provencher 均指出，多智能体主要带来速度而非质量提升，协调成本高，多数场景不划算。
正文节选
AI agent teams waste massive tokens for barely measurable quality gains, research finds AI agent teams deliver almost no better results than single agents, research finds. Evals company Vals AI tested GPT-6 Sol and Claude Opus 5.5 on the "Vibe Code Bench," both solo and as teams, at two reasoning levels: medium and maximum reasoning effort. The teams cost between 1.8x and 5.1x more than single agents. Out of four comparisons between teams and solo agents, only one showed a statistically signific