Nvidia’s Vera Rubin NVL72 trains models with 75% fewer GPUs than Grace Blackwell

Nvidia's Vera Rubin NVL72, packing 72 Rubin GPUs and 36 Vera CPUs into a single NVLink domain, can train a 10-trillion-parameter mixture-of-experts model on 100 trillion tokens in one month using about 75% fewer GPUs than the Grace Blackwell NVL72, Nvidia announced. Nvidia projects inference on the Vera Rubin NVL72 will cost one-tenth the price per million tokens versus Blackwell racks, with MLPerf Inference v6.1 (September 2026) recording up to 3.7 times higher throughput than the GB300 NVL72 on the Qwen3-VL task and Artificial Analysis showing roughly 5.7 times higher throughput per megawatt on agentic AI workloads. Production shipments ramped in fall 2026 with support from over 80 MGX partners, including Google Cloud, which anticipates scaling to as many as 960,000 Rubin GPUs by November, and Microsoft Azure.

NVIDIA official newsroom nvidianews.nvidia.com/bios/jensen-huang Nvidia’s Vera Rubin NVL72 trains models with 75% fewer GPUs than Grace Blackwell Nvidia's next-generation supercomputer promises to slash hardware requirements and cut inference costs to one-tenth of its predecessor's price per million tokens Nvidia https://cryptobriefing.com/markets/nvidia/ has a habit of making its last chip look obsolete before most customers have finished unpacking it. The Vera Rubin NVL72 is the latest case in point. The new system, which packs 72 Rubin GPUs and 36 Vera CPUs into a single NVLink domain, can train a 10-trillion-parameter mixture-of-experts model on 100 trillion tokens in one month using about 75% fewer GPUs than the Grace Blackwell NVL72 configuration. What the Vera Rubin NVL72 actually is The system is the direct successor to Nvidia’s Blackwell NVL72 racks, which shipped in GB200 and GB300 variants. Each GPU in the Vera Rubin NVL72 carries 288 GB of HBM4 memory. The architecture also introduces new interconnectivity designs and updated precision formats, both aimed at squeezing more useful work out of every watt consumed. The cost reduction on the inference side is equally striking. Nvidia projects that running inference on the Vera Rubin NVL72 will cost one-tenth the price per million tokens compared to Blackwell racks. AI, tech, and the markets they move—in one daily briefing. Daily. Free. Join 34,000+ readers across crypto, finance, and policy. Third-party benchmarks back up at least part of that story. MLPerf Inference v6.1, released in September 2026, recorded throughput improvements of up to 3.7 times higher than the GB300 NVL72 on the Qwen3-VL task. Separately, Artificial Analysis benchmarks showed approximately 5.7 times higher throughput per megawatt than the GB300 NVL72 on agentic AI workloads. Who is buying it and what they plan to do with it Production shipments began ramping in fall 2026, with support from over 80 MGX partners. The two most consequential names on that list are Google Cloud and Microsoft https://cryptobriefing.com/markets/microsoft/ Azure. Google Cloud anticipates scaling to as many as 960,000 Rubin GPUs by November. Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our Editorial Policy https://cryptobriefing.com/editorial-policy/ .