Vera Rubin 10x token throughput per MW vs GB200 on DeepSeek R1
Company: CoreWeave
The claim, verbatim
CoreWeave published measured silicon performance showing 10x the token throughput per megawatt over NVIDIA GB200 NVL72 on DeepSeek R1 at matched interactivity.
Source (primary)
CoreWeave Delivers NVIDIA Vera Rubin NVL72 at Scale - CoreWeave (-, news_article)
View cached copy (2026-10-05)Live source ↗Archive.org ↗
Quote: “showing 10 times the token throughput per megawatt over NVIDIA GB200 NVL72 on the DeepSeek R1 reasoning model at matched interactivity”
How we checked this
Checked on October 6, 2026. The cited source supports every part of this claim.
The release states that CoreWeave published measured silicon numbers showing 10x token throughput per megawatt over GB200 NVL72 on DeepSeek R1 at matched interactivity.
Confirmed in the source:
- CoreWeave published measured silicon performance numbers on the Vera Rubin platform
- The numbers show 10x the token throughput per megawatt over NVIDIA GB200 NVL72
- The result was on the DeepSeek R1 reasoning model at matched interactivity
What we did: Read our cached copy of the publisher (https://www.coreweave.com/news/coreweave-delivers-nvidia-vera-rubin-nvl72-performance-at-production-scale-starting-with-cognition) in full (13,996 characters, retrieved October 5, 2026) and checked each assertion in the claim against it.
Additional evidence
confirms CoreWeave Delivers NVIDIA Vera Rubin NVL72 at Scale - CoreWeave
Quote: “showing 10 times the token throughput per megawatt over NVIDIA GB200 NVL72 on the DeepSeek R1 reasoning model at matched interactivity”