论文40

混合模型团队基准测试表现更优

原标题:The results hold on both benchmarks. Every mixed team beat four copies of its lead model: +0.008 to +0.125 on Terminal-Bench 4.1, +0.015 ...

Perplexity@perplexity_ai发布时间 (北京时间)

结果在两个基准测试上都成立。每个混合团队都击败了其主导模型的四个副本:在 Terminal-Bench 4.1 上提升 +0.008 至 +0.125,在 WANDR 上提升 +0.015 至 +0.064。 每个基准上排名最高的混合团队成本也低于其主导模型的四个副本:每次运行 $10.31 对约 $13,以及 $54.50 对 $66.41。

中文译文尚未完成,先为你展示原文。

The results hold on both benchmarks. Every mixed team beat four copies of its lead model: +0.008 to +0.125 on Terminal-Bench 4.1, +0.015 to +0.064 on WANDR.

The top mixed team on each also cost less than four copies of its lead: $10.31 vs. ~$13 per run, and $54.50 vs. $66.41.

Agent 智能体论文研究评测基准#论文/研究#Agent#评测/基准

同一件事的其他报道2 篇