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

NVIDIA announced that its Vera Rubin NVL72 system made its MLPerf Inference v6.1 debut with leading performance, delivering up to 3.7x higher throughput than GB300 NVL72 on the Qwen3-VL benchmark. The company also reported 99% scaling efficiency for a 288-GPU GB300 NVL72 submission and up to 1.6x performance gains from software optimizations over v6.0. The results highlight performance, scaling efficiency, and software velocity as key factors for AI inference economics.