事實(公司自述/申報)NVIDIA(SEC Form 8-K 附件 99.1 財報新聞稿)2026-08-26
Vera Rubin 已經全面量產;機櫃已經在 CoreWeave、Google Cloud、微軟 Azure、甲骨文雲端、Nebius 等夥伴那裡運轉。
原文
“Vera Rubin, now in full production, was built to power exactly this moment. / Announced the NVIDIA Vera Rubin platform is ramping into full production with racks running at partners including CoreWeave, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure and Nebius.”
“Hyperscale 48,710 43,050 24,168 … Hyperscale revenue more than doubled from a year ago and increased 13% sequentially on the strength of Blackwell Ultra.”
位置:10-Q 第 27–28 頁(MD&A:Revenue by Market Platform 表與其下說明)・原件 ↗
“Rubin integrates up to 288 GB of HBM4 memory, driven by dedicated HBM controllers and 12-Hi stacks, to deliver up to 22 TB/s of peak bandwidth. / HBM4 doubles interface width relative to HBM3e … this subsystem provides up to 22 TB/s of memory bandwidth: a 2.8x increase over Blackwell and Blackwell …”