NVIDIA Vera Rubin NVL72 在 MLPerf Training v6.1 中达到 Blackwell Ultra 最高 2 倍性能
NVIDIA Vera Rubin NVL72 Achieves up to 2x Blackwell Ultra Performance in MLPerf Training v6.1
NVIDIA Vera Rubin 架构首次亮相 MLPerf Training v6.1,Vera Rubin NVL72 平台取得最快训练时间,性能最高达 Blackwell Ultra 的 2 倍。NVIDIA 表示其 CUDA 软件栈在首次登场时已相当成熟,缩短训练时间对快速迭代模型和智能体至关重要。
As AI agents tackle increasingly complex workflows, reducing time-to-train is critical for rapidly iterating on both models and the agents they power. MLPerf...
As AI agents tackle increasingly complex workflows, reducing time-to-train is critical for rapidly iterating on both models and the agents they power. MLPerf Training v6.1 marks the debut of the NVIDIA Vera Rubin architecture, and these early results demonstrate that its CUDA software stack is already highly mature. In its first appearance, the Vera Rubin NVL72 platform delivered the fastest time…
来源:NVIDIA Technical Blog · developer.nvidia.com