The Frontier · Tags
#mlperf
2 published articles
Record: MLPerf Inference v6.1 preview table for Vera Rubin NVL72 vs GB300 NVL72
Evidence document for the 16 September 2026 MLPerf Inference v6.1 round, focused on NVIDIA's Vera Rubin NVL72 preview claims and the supporting MLCommons artifacts.
Vera Rubin NVL72's first MLPerf Inference v6.1 numbers: up to 3.7x vs GB300 on Qwen3-VL
NVIDIA's 16 September 2026 preview submission puts Vera Rubin NVL72 on DeepSeek-R1 and Qwen3-VL. Distinct from the earlier Vera Rubin production dispatch: this is the benchmark stack.