Nvidia + Google Cloud ‘AI factories’

Nvidia and Google Cloud today unveiled a packaged offering that lets enterprises build “AI factories” using Vera Rubin–powered A5X instances and Google Distributed Cloud. (blogs.nvidia.com) The companies announced the move at Google Cloud Next in Las Vegas on April 22, 2026, during a session describing expanded co‑engineering between the two firms. (blogs.nvidia.com) Nvidia says A5X uses Vera Rubin NVL72 rack‑scale systems with ConnectX‑9 SuperNICs and Google’s Virgo networking, scaling to 80,000 Rubin GPUs per single‑site cluster and up to 960,000 Rubin GPUs across multisite clusters. (blogs.nvidia.com) The bundle also previews Google’s Gemini model running on Google Distributed Cloud, confidential virtual machines backed by Nvidia Blackwell and Blackwell Ultra GPUs, and integration of Nvidia Nemotron models and the NeMo framework on Gemini Enterprise Agent Platform. (blogs.nvidia.com) The package pairs hyperscale Rubin capacity with distributed deployment and confidential GPU options, giving customers hardware choices for on‑premises, edge and multisite cloud placement. (blogs.nvidia.com) An “AI factory” here means a tightly co‑designed hardware and software stack — chips, racks, networking and model tooling — built so organizations can train, tune and serve large or agentic models across multiple locations. (nvidianews.nvidia.com) Nvidia touts up to 10x lower inference cost per token and 10x higher token throughput per megawatt versus prior generations when A5X runs Vera Rubin NVL72 hardware, according to the company’s blog post. (blogs.nvidia.com) Google is still developing its own TPU lineup (TPU v8 family and new TPU variants announced this year) even as it expands Nvidia options, illustrating parallel bets on in‑house silicon and Nvidia’s Rubin/Blackwell stack. (eweek.com) “At Google Cloud, we believe the next decade of AI will be shaped by customers’ ability to run their most demanding workloads on a truly integrated, AI‑optimized infrastructure stack,” said Mark Lohmeyer, vice president and general manager of AI and computing infrastructure at Google Cloud. (blogs.nvidia.com)