一句话：首次系统实证三种轮次控制策略，提出按需延长的动态轮次策略，在保持解决率的同时大幅降低成本。
LLM 驱动的编码智能体通过多轮（turn）迭代完成软件工程任务，但 token 随轮次近似二次增长、顶尖模型价格高昂、真实任务所需轮次多、且智能体常有低效动作，导致部署成本高且不可预测。已有研究多关注单轮优化，而总轮数的策略性控制尚缺乏探索。本文在 SWE-bench 上使用 Claude 4 Sonnet、Gemini 2.5 Pro 与 GPT 4.1 三种先进模型，系统比较三种策略：无限制基线、带提醒的固定轮次上限、以及按需授予延长的动态轮次策略。结果显示：无限制设置下性能、成本与轮次效率之间存在根本性权衡，没有单一模型全面占优；固定上限取基线第 75 百分位是“甜点”，可降低成本 24%–68% 而对解决率影响很小；动态策略持续优于固定上限，在解决率相当或更好的情况下再额外降本 12%–24%，因为它只把资源分配给真正需要的任务。该工作给出了简单易行的成本—效果平衡指南。
LLM-powered coding agents, which operate in iterative loops (turns) to solve software engineering tasks, are becoming increasingly powerful. However, their practical deployment is hindered by significant and unpredictable costs. This challenge arises from a combination of factors: quadratically growing token counts with each turn, the high price of state-of-the-art models, the large number of turns required for real-world tasks, and the tendency of agents to take inefficient or unnecessary actions. While existing research focuses on optimizing individual turns, the strategic control of the total number of turns remains an underexplored area for managing agent performance and cost. To address this gap, we conduct a comprehensive empirical study on the SWE-bench benchmark using three state-of-the-art models (Claude 4 Sonnet, Gemini 2.5 Pro, and GPT 4.1). We systematically evaluate the impact of three distinct turn-control strategies: an unrestricted baseline, a fixed-turn limit with reminders, and a novel dynamic-turn strategy that grants extensions on-demand. Our findings first reveal a fundamental trade-off in the unrestricted setting, where no single model excels across performance, cost, and turn efficiency. We then show that a fixed-turn limit, specifically at the 75th percentile of the baseline, serves as a "sweet spot", substantially reducing costs (by 24%-68%) with minimal impact on solve rates. Most significantly, our proposed dynamic-turn strategy consistently outperforms fixed-limit approaches, achieving comparable or better solve rates while further reducing costs by an additional 12%-24% by intelligently allocating resources only to tasks that need them. This work provides the first systematic analysis of turn-control strategies, offering simple yet effective guidelines for developers to balance cost and efficacy. We demonstrate that dynamic resource allocation is a superior, easy-to-implement approach for deploying powerful yet economically viable coding agents.