Google Cloud plans NVIDIA Vera Rubin A5X instances next month
Original title: Google Cloud and NVIDIA Announce General Availability of NVIDIA Vera Rubin Infrastructure
A5X pairs Vera Rubin NVL72 with ConnectX-9 networking, and Google plans capacity of up to 960,000 Rubin GPUs.
NVIDIA and Google Cloud said general availability of Vera Rubin-based A5X instances is coming next month, with later expansion planned in Europe. A5X combines NVIDIA Vera Rubin NVL72, ConnectX-9 SuperNICs, the Virgo network and fourth-generation liquid cooling for training, inference and reinforcement learning. Ineffable Intelligence was the first customer; WPP demonstrated creative agents using Nemotron and OpenShell. Google Cloud also described rack-scale Confidential Computing plans and general availability of Confidential Computing on G4 VMs.
Unconfirmed: Instance availability next month, European expansion and the 960,000-GPU figure are stated plans, not current capacity.
- Google Cloud plans as many as 960,000 Rubin GPUs across sites.
- Ineffable Intelligence, founded by David Silver, accessed A5X first.
- Confidential Computing is planned on bare-metal A4X Max instances.
- Confidential Computing is generally available on G4 VMs with RTX PRO 6000 Blackwell GPUs.
- WPP's initial tests ran multi-node workloads on one A5X instance.
- Vendors
- NVIDIAGoogle Cloud
- Organizations
- Ineffable IntelligenceWPP
- AI models
- NVIDIA Nemotron
From the original article
2 min read at blogs.nvidia.com
General availability of the instances is coming next month, with planned expansion of availability to Europe — bringing an integrated platform for training, inference and large-scale reinforcement learning to cloud customers.
By Uttara Kumar · blogs.nvidia.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.