NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut

NVIDIA’s Vera Rubin NVL72 architecture achieved top performance results in the latest MLPerf Inference v6.1 industry benchmarks. These figures highlight how the new hardware improves token generation speed and infrastructure scalability, which are critical metrics for optimizing the economics of large-scale AI deployments.