petergostev· @petergostev · X·· 5 天前AI 评分
Mistral Large 4 与 Astra 训练规模对比
Mistral Large 4 用了 4,000 块 GPU 训练，而 Astra 用了 100,000 块。 我们看看这只小猫能做到什么，但差距相当悬殊。
Mistral Large 4 was trained on 4,000 GPUs and Astra was trained on 100,000 of them.
We'll see what the kitten can pull off, but the disparity is pretty stark
来源：petergostev · x.com