教程40
Together AI 如何做智能体评测
原标题:Here's how we do agent evals at Together AI:
我们在 Together AI 是这样做智能体评测的: → 定义约 10 个代表性任务 → 在沙箱中运行智能体(Claude Code) → 每个任务测试 3 种配置(baseline、MCP、skills) → 分析轨迹并改进文档 + API 智能体任务完成率从 67% 提升到 96%!
中文译文尚未完成,先为你展示原文。
Here's how we do agent evals at Together AI:
→ Define ~10 representative tasks
→ Run agents (Claude Code) in sandboxes
→ Test 3 configs (baseline, MCP, skills) for each task
→ Analyze trajectories and improve docs + APIs
Agent task completion improved from 67% to 96%!
https://x.com/i/article/2122371457288200192查看原帖
