NVIDIA Blog· Zhihan Jiang·· 15 d ago
NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut
NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut
AI summary
NVIDIA submitted preview results for Vera Rubin NVL72 to MLPerf Inference v6.1 for the first time, achieving throughput up to 3.7 times that of GB300 NVL72 on Qwen3-VL and up to 2.5 times on DeepSeek-R1.
Selection record
Threshold 60Official, first-handFirst 62Second 62
AdmittedSum of both 124 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：NVIDIA AI推理平台MLPerf性能结果
- Why it was chosen
- The throughput and scaling-efficiency figures from its first MLPerf inference submission inform assessment of value-for-money trends in inference infrastructure.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: NVIDIA Blog · blogs.nvidia.com