分析4300次编程智能体使用，揭示其高效服务优化方向
TraceLab: Characterizing Coding Agent Workloads for LLM Serving
- 收集4300个会话的真实使用数据，涵盖35万次LLM步骤和43万次工具调用
- 发现工作负载具长循环、长上下文短输出、工具调用分布极不均衡等特征
- 适合研究大模型服务优化的工程师与系统设计者参考
编程智能体正成为智能体型大模型的主要应用，但其高效服务仍面临挑战。这一挑战的突破需依赖对真实工作负载模式的理解，然而相关数据极为稀缺。现有公开日志与基准测试无法覆盖多智能体、多模型家族在实际服务中的日常使用情况。为此，我们收集并发布了一个约4,300个编程智能体会话的追踪数据，包含约35万次LLM步骤和43万次工具调用，来自我们日常使用Claude Code和Codex的真实场景。分析显示，编程智能体工作负载具有长自主循环、长上下文伴随短输出、工具调用多样且呈重尾分布、前缀缓存命中率高但不完美等特点。这些发现为服务优化提供了具体方向，包括降低工具调用开销、基于追加长度的预填充优化、语义感知的工具延迟预测，以及针对人类操作节奏间隙的KV缓存管理改进。数据集、追踪管道与分析代码已开源至https://github.com/uw-syfi/TraceLab.git，项目主页为https://tracelab.cs.washington.edu。
原文摘要 · Abstract (English)
Coding agents are rapidly becoming a major application of agentic LLMs, but serving them efficiently remains challenging. Progress on this challenge requires understanding real workload patterns, yet the data needed for such analysis is largely absent. Existing public traces and benchmarks do not capture real, day-to-day coding-agent usage across multiple agents and model families for serving-system analysis. To help fill this gap, we collect and release a trace of roughly 4,300 coding-agent sessions, containing about 350,000 LLM steps and 430,000 tool calls from our own day-to-day use of Claude Code and Codex. Our analysis shows that coding-agent workloads feature long autonomous loops, long contexts with short outputs, diverse and heavily-tailed tool calls, and high but imperfect prefix cache hit rates. These findings point to concrete opportunities for optimizing serving, including lower-overhead tool calling, append-length-aware prefill, semantic-aware tool-latency prediction, and improved KV-cache management around human-paced gaps. We release the dataset, trace collection pipeline, and analysis code at https://github.com/uw-syfi/TraceLab.git the project website is https://tracelab.cs.washington.edu.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。