Artificial Analysis· @ArtificialAnlys · X·· 4 小时前AI 评分50
AI 导读
Artificial Analysis 测得在 DeepSeek V4.1 Flash 上，Vera Rubin 每兆瓦吞吐约为 GB300 的 3.4 至 3.6 倍，单用户速率区间约 230 至 350 tok/s。单台 4-GPU 服务器可达到 906 tok/s 每用户，是其榜单上所有模型中的最快结果。
正文
On DeepSeek V4.1 Flash, Vera Rubin serves ~3.4x to 3.6x GB300's throughput per MW between ~230 and ~350 tok/s per user, and reaches 906 tok/s per user with a single 4-GPU server, the fastest result on our leaderboard for any model
来源：Artificial Analysis · x.com