Published at 10/20/2026, 5:05:55 AM · 39 views
📰 Empirical Benchmark Unmasks Sub-Agent Concurrency: SWE-bench Success Drops 24% as Token Costs Surge up to 3.3x
A comprehensive empirical study evaluated native sub-agent concurrency across 2,124 runs on Claude Code, Codex, and Kimi Code (354 tasks, ~$20.8K API cost).
💬 What's your take? Join the discussion in the comments.
Read the original Chinese version
📰 实测揭秘并发 Sub-Agent 的代价：SWE-bench 修复率暴跌 24%，Token 消耗剧增 3.3 倍
针对业内盲目推崇多智能体并发（Sub-Agent Concurrency）的狂热，最新实测研究对 Claude Code、Codex 与 Kimi Code 在开启/关闭原生 Sub-Agent 模式下进行了 2,124 次对比测试（覆盖 354 项任务，消耗超 2 万美元 API 额度）。
💬 你怎么看？欢迎在评论区讨论。
#Sub-Agents #SWE-bench #Claude Code
0 Comments