跳到正文
Nathan Lambert· @natolambert · X·· 3 小时前AI 评分23
AI 导读

我觉得有必要同时为 AAv5 指数和开放模型辩护。基准测试比大多数人以为的更混乱、更主观、也更棘手。

正文

I feel a need to defend the AAv5 index and open models at the same time. Benchmarking is messier, more opinionated, and trickier than most people give credit to.

来源:Nathan Lambert · x.com