AgentOpt v0.1 技术报告：基于LLM的代理的客户端优化
AgentOpt v0.1 Technical Report: Client-Side Optimization for LLM-Based Agent
AI总结：
本文提出AgentOpt框架，用于解决LLM代理客户端资源分配问题，通过多种搜索算法优化模型选择和资源分配，提升效率与效果。
摘要：
AI代理越来越多地应用于现实世界，包括Manus、OpenClaw和编码代理等系统。现有研究主要关注服务器端效率，提出缓存、推测执行、流量调度和负载均衡等方法以降低服务代理工作负载的成本。然而，随着用户越来越多地通过组合本地工具、远程API和多样模型来构建代理，客户端侧的优化问题同样重要。客户端优化探讨开发者如何在应用特定的质量、成本和延迟约束下分配可用资源，包括模型选择、本地工具和API预算等。由于这些目标依赖于任务和部署设置，无法仅由服务器端系统决定。我们引入AgentOpt，首个框架无关的Python包，用于客户端代理优化。首先研究模型选择，多步骤代理管道中的高影响优化杠杆。给定一个管道和一个小的评估集，目标是找到最经济的模型分配方案。在实践中，匹配精度下，最佳与最差模型组合的成本差距可达13-32倍。为高效探索指数增长的组合空间，AgentOpt实现了十种搜索算法，包括UCB-E、UCB-E与低秩分解、Arm消除、Epsilon-LUCB、阈值成功消除和贝叶斯优化。在四个基准测试中，UCB-E在减少62-76%的评估预算的同时恢复了接近最优的准确性。代码和基准结果可在https://agentoptimizer.github.io/agentopt/上获取。
英文摘要：
AI agents are increasingly deployed in real-world applications, including systems such as Manus, OpenClaw, and coding agents. Existing research has primarily focused on server-side efficiency, proposing methods such as caching, speculative execution, traffic scheduling, and load balancing to reduce the cost of serving agentic workloads. However, as users increasingly construct agents by composing local tools, remote APIs, and diverse models, an equally important optimization problem arises on the client side. Client-side optimization asks how developers should allocate the resources available to them, including model choice, local tools, and API budget across pipeline stages, subject to application-specific quality, cost, and latency constraints. Because these objectives depend on the task and deployment setting, they cannot be determined by server-side systems alone. We introduce AgentOpt, the first framework-agnostic Python package for client-side agent optimization. We first study model selection, a high-impact optimization lever in multi-step agent pipelines. Given a pipeline and a small evaluation set, the goal is to find the most cost-effective assignment of models to pipeline roles. This problem is consequential in practice: at matched accuracy, the cost gap between the best and worst model combinations can reach 13-32x in our experiments. To efficiently explore the exponentially growing combination space, AgentOpt implements ten search algorithms, including UCB-E, UCB-E with Low-Rank Factorization, Arm Elimination, Epsilon-LUCB, Threshold Successive Elimination, and Bayesian Optimization. Across four benchmarks, UCB-E recovers near-optimal accuracy while reducing evaluation budget by 62-76\% relative to brute-force search. Code and benchmark results available at https://agentoptimizer.github.io/agentopt/.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
