Cohere 论文：编码智能体团队中让廉价模型只做规划可降本 70%
New Cohere paper finds that a team of coding agents gets the most from extra agents when they only research and plan, and cheap models can do that job for a str
Cohere 一篇新论文发现，编码智能体团队中让额外智能体只负责研究和规划，能获得最大收益，且廉价模型即可胜任。研究者在 ProgramBench 上让智能体仅凭可执行文件和文档重建整个程序，只做规划的 worker 达到与全阶段 worker 相同的奖励，成本减半；由 GPT-5.6 Sol 担任 lead、搭配 16 个廉价 Luna 规划者，与全 Sol 团队相差 1.4 分以内，成本低 70%。作者建议把最强模型放在 lead 位置，廉价模型只用于规划，因为廉价编码模型会拉低奖励。
正文
当前显示原文，中文全文译文尚未就绪。
New Cohere paper finds that a team of coding agents gets the most from extra agents when they only research and plan, and cheap models can do that job for a strong lead at far lower cost.
A team of coding agents beats 1 agent working alone, but costs 3 to 7 times as much. More agents can also mean duplicate work and extra code to merge.
The researchers built a lead agent that hands subtasks to workers, then varied the workers' role, model and count. On ProgramBench, agents rebuild a whole program from just a working executable and its docs.
Workers that only plan matched the reward of workers used in every stage, at half the cost. A strong GPT-5.6 Sol lead with 16 cheap Luna planners came within 1.4 points of an all-Sol team at 70% lower cost.
On tasks like this, put your strongest model in the lead and use cheap models only for planning, since cheap coders lowered reward.