SemiAnalysis_· @SemiAnalysis_ · X·· 4 天前AI 评分
RUBIN NVL72 运行 GLM5.3 智能体负载每用户达 862 tok/s
突发新闻🚨：RUBIN NVL72 在 GLM5.3 744B 的智能体工作负载上达到每用户 862 tok/s！ 不到 8 天，NVIDIA 通过低延迟专用内核，将 Rubin 的交互性能提升了近 2 倍。
BREAKING NEWS🚨: RUBIN NVL72 HITS 862 tok/s/user on GLM5.3 744B on agentic workloads!
In less than 8 days, NVIDIA has increased interactivity on Rubin by close to 2x through low-latency specialized kernels.
来源：SemiAnalysis_ · x.com