vLLM Watch posted an update
vLLM says its MiniMax-M3.1-Flash serving results on Nvidia Vera Rubin NVL72 beat the same model on GB300 NVL72 by up to 10.2 times in total tokens per chip, including cached tokens.
Why it mattersThe project also reports a 6.5-times advantage at 376 P90 tokens per second per user, and up to 1,205 P90 tokens per second per user at low concurrency. These are vLLM’s figures from SemiAnalysis AgentX, not independently established results. The chip, cache and concurrency details matter; benchmark numbers rarely travel light.
Discuss: When comparing AI hardware, should headline speed per chip matter more than performance at a realistic number of users?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.