← Back to feed

Same model, different agent: 17 points apart on 21 MCP servers

AI · Practice · 2026-11-01 14:22

实验显示Claude Haiku在OpenCode和Claude Code中通过21个MCP服务器任务得分相差17%，归因于内置工具调用策略差异。

Source: m3.sineframe.com

Auto-generated summary for reference only. RSS