FrugalEvo：让强模型出策略、便宜模型写代码，圆 packing 成本降到 0.55 美元
FrugalEvo: Towards Cost-Aware LLM-Guided Program Evolution
花 0.55 美元干出别人 50 美元的效果，强模型出思路、便宜模型搬砖的省钱组合拳，搞优化任务的可以看看。
FrugalEvo 是一个成本感知的 LLM 进化框架：贵的 LLM（如 GPT-5.6 Terra/Luna、GLM-5.3）负责探索解法策略，便宜的 LLM 负责实现和迭代代码。它还设计了前缀共享的缓存复用机制来降低 token 开销，并提出按成本预算衡量效果的 BA-AUC 指标。在 10 个数学和系统优化任务上，FrugalEvo 匹配或超过 OpenEvolve、ShinkaEvolve、AdaEvolve、EvoX 等基线，并在其中 9 个任务上 BA-AUC 更高。在 circle packing 任务上，它用 GPT-5.6 系列只花 1.68 美元、用 GLM-5.3 只花 0.55 美元就达到新的 SOTA，而 CORAL、SwarmResearch 等多智能体基线平均要花约 50 美元。
FrugalEvo: Towards Cost-Aware LLM-Guided Program Evolution
LLM-guided evolutionary methods, such as AlphaEvolve, have emerged as powerful approaches for challenging computational optimization problems, such as circle packing. However, prior work typically optimizes performance gain over a fixed number of iterations. We argue that practical optimization should maximize gain per unit cost. To this end, we propose FrugalEvo, a cost-aware evolutionary framework where a stronger, higher-cost LLM explores solution strategies, and a cheaper LLM implements them and iteratively refines the resulting code. We also design a cache-efficient evolution process, where our harness and prompts maximize the sharing of prefixes across different evolution steps, to improve cache reuse. To measure solution quality throughout a fixed cost budget, we introduce Budget-Aware Area Under the Curve (BA-AUC), defined as the area under the best-so-far evaluation score curve over cumulative LLM cost, up to the budget. Across 10 mathematical and systems optimization tasks, FrugalEvo matches or surpasses state-of-the-art baselines, including OpenEvolve, ShinkaEvolve, AdaEvolve, and EvoX, in final solution quality and achieves higher BA-AUC on 9 tasks. It also achieves higher average performance than these baselines on 10 algorithmic optimization tasks from ALE-Bench-Lite. Notably, on circle packing, FrugalEvo achieves new state-of-the-art performance with GPT-5.6 Terra and Luna for only 1.68 USD and with GLM-5.3 and its Flash variant for only 0.55 USD, matching or surpassing all baselines, including multi-agent methods such as CORAL and SwarmResearch, which cost approximately 50 USD on average.
