用 Jev 构建代码库分类器
开发者分享用 Jev 构建代码库分类器,认为这可能解决智能体生成过度工程化代码的问题,并询问下一步测试方向。
Jev 被用于判断哪个模型是 AGI,速度提升 20–200 倍,成本降低 40–400 倍,可让 LLM-as-a-judge 变得极快且近乎免费。
Jev has spoken. It picked which model is AGI. 20–200x faster. 40–400x cheaper. This could make things like LLM-as-a-judge insanely fast and nearly free. (I tried a bunch of prompts and still didn’t burn through $0.10.)