GA4GC: Greener Agent for Greener Code via Multi-Objective Configuration Optimization
This addresses the problem of high environmental costs and inefficiency in industrial deployment of coding agents, though it is incremental as it builds on existing agent optimization methods.
The paper tackled the sustainability and scalability challenges of LLM-powered coding agents by introducing GA4GC, a framework that optimizes trade-offs between agent runtime and code performance, achieving up to 135x hypervolume improvement and reducing runtime by 37.7% while improving correctness.
Coding agents powered by LLMs face critical sustainability and scalability challenges in industrial deployment, with single runs consuming over 100k tokens and incurring environmental costs that may exceed optimization benefits. This paper introduces GA4GC, the first framework to systematically optimize coding agent runtime (greener agent) and code performance (greener code) trade-offs by discovering Pareto-optimal agent hyperparameters and prompt templates. Evaluation on the SWE-Perf benchmark demonstrates up to 135x hypervolume improvement, reducing agent runtime by 37.7% while improving correctness. Our findings establish temperature as the most critical hyperparameter, and provide actionable strategies to balance agent sustainability with code optimization effectiveness in industrial deployment.