NVIDIA Vera Rubin NVL72 posts leading MLPerf Inference v6.1 results
NVIDIA released preview results showing Vera Rubin NVL72 achieves up to 3.7x higher throughput than GB300 NVL72 on Qwen3-VL and 2.5x on DeepSeek-R1 in the MLPerf Inference v6.1 suite.