Nscale Raises $2 Billion in Series C — the Largest in European History
See More
See More
Unlock gigascale AI training and inference with NVIDIA DGX Vera Rubin NVL72, a turnkey rack-scale AI infrastructure platform built for next-generation AI factories, reasoning models, simulation, and enterprise-scale inference.
NVFP4 inference performance
Built to accelerate large-scale AI inference workloads.
NVFP4 training performance
Engineered for frontier-scale model training and post-training.
Total fast memory
Massive memory capacity for reasoning, simulation, and large AI workloads.
Unified by sixth-generation NVLink
Connected as one massive GPU fabric for rack-scale performance.
DGX Vera Rubin NVL72 brings together 72 NVIDIA Rubin GPUs and 36 NVIDIA Vera CPUs in a 100% liquid-cooled rack-scale system. Designed for AI factories, it delivers the compute, memory, networking, and integrated software needed to move from experimentation to production at gigascale.
