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
Vera Rubin serves DeepSeek V4.1 Flash 3.4–3.6x GB300 throughput per MW
AIArtificial Analysis reports Vera Rubin delivers about 3.4x to 3.6x GB300's throughput per MW on DeepSeek V4.1 Flash, at roughly 230 to 350 tok/s per user. A single 4-GPU Vera Rubin server reaches 906 tok/s per user, the fastest result on its leaderboard for any model.
Source details
Artificial Analysis · x.comPublished
Original title: 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 t...