smevals - a small eval suite for evaluating models, prompts, and harnesses

smevals——用于评估模型、提示词和测试框架的小型评估套件

来源: Simon Willison | 时间: 2026-07-31T21:15:23+00:00

smevals - a small eval suite for evaluating models, prompts, and harnesses

smevals - 一个用于评估模型、提示词及测试框架的小型评估套件

I've been working with Jesse Vincent's Prime Radiant applied AI research lab building out this evals framework to help answer questions about the capabilities of different models.

⋯ 继续阅读请登录会员 ⋯