Liucong Model Eval

初始化并执行可追溯的模型对照测评,支持权威视觉 benchmark 子集、用户自有题库和前后端真实任务;固定题面、隔离工具、保留首轮产物并验收。用于从零搭建测评环境、按个人题库跑测、同题比较模型或整理测评报告。

liucongg Updated

File contents

liucongg/liucong-skills commit 6e379676ba

Frequently asked questions

npx get-skill add liucongg/liucong-model-eval