AI 导读
执行器越便宜,提升幅度越大。 在相同的评测中,Haiku 搭配 Opus 顾问在 BrowseComp 上达到 41.2%,是其单独运行 19.7% 得分的两倍多,而成本仅为每个 token 都跑前沿模型的一小部分。
正文
The cheaper your executor, the bigger the jump.
In the same evals, Haiku with an Opus advisor hit 41.2% on BrowseComp, more than double its 19.7% solo score, for a fraction of what running the frontier model on every token would cost.
来源:@OpenRouter · x.com