More with Less: An Empirical Study of Turn-Control Strategies for Efficient Coding Agents

TL;DR

This paper tackles the high operational cost of LLM-powered coding agents by systematically studying turn-control strategies. It presents an empirical evaluation of unlimited turns, fixed-turn limits with reminders, and a novel dynamic-turn allocation across three leading closed-source models on a realistic SWE-bench Verified subset. The findings show a clear cost-performance trade-off: a fixed-turn sweet spot at the 75th percentile substantially reduces costs with minimal impact on solve rates, while the dynamic-turn approach consistently outperforms fixed limits by allocating turns only where needed. The work offers practical guidelines and demonstrates that dynamic resource allocation is a superior, easy-to-implement strategy for deploying powerful yet economically viable coding agents.

Abstract

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 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 SWE-bench using three state-of-the-art models and 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, the 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.

Figures (2)

• Figure 1: Number of Turns Needed by Different Agents
• Figure 2: The workflow of the autonomous agent, illustrating the core loop of agent and the turn-control mechanism