SpaceXAI Picks NVIDIA Vera CPU to Power Agentic AI on Earth and in Orbit

SpaceXAI will deploy NVIDIA Vera CPUs to accelerate its next generation of agentic AI applications, NVIDIA announced. The move brings NVIDIA's first CPU built for AI agents into one of the largest AI deployments, and SpaceXAI plans to extend the same accelerated computing architecture into space aboard a planned Starmind AI satellite.

Agentic AI applications increasingly depend on CPUs to orchestrate tools, execute code, process data and run simulations between model calls. SpaceXAI will use Vera to accelerate these workloads, aiming to help AI agents act faster while keeping GPUs fully utilized. The company plans to expand the AI infrastructure behind Grok on the NVIDIA Vera Rubin platform as it scales toward gigawatts of computing capacity.

NVIDIA Vera is described as the first CPU built for AI agents, designed to accelerate the CPU-intensive work surrounding model inference, including tool use, code execution, data processing, orchestration and simulation. Vera features 88 NVIDIA-designed Olympus cores, NVIDIA Spatial Multithreading technology and high-bandwidth LPDDR5X memory delivering up to 1.2TB/s of bandwidth. It enables up to 1.8 times faster task completion compared with x86 CPUs across workloads including agentic AI, reinforcement learning and data processing.

For its terrestrial infrastructure, SpaceXAI will build on NVIDIA Vera Rubin, a platform codesigned across compute, networking and software. It brings together NVIDIA accelerated computing, NVLink interconnect technology, Spectrum-X Ethernet networking, BlueField data processing and NVIDIA software in an integrated architecture intended to deliver performance and energy efficiency at scale while driving down cost per token. As SpaceXAI expands toward gigawatts of capacity, Vera Rubin provides a common architecture to scale its next generation of AI factories.

SpaceXAI's planned first-generation Starmind AI satellite will be based on an optimized NVIDIA Vera Rubin NVL72 rack-scale system, extending the same architecture powering its AI factories on Earth into orbit. The company is developing AI computing infrastructure for space, where power, thermal management, bandwidth, reliability and physical integration impose dramatically different constraints from conventional data centers. NVIDIA and SpaceXAI are working to adapt that foundation to orbital requirements while preserving a common architecture and software ecosystem.

"Agentic AI requires a new kind of computing system, one built not only to generate answers, but to take action," said Ian Buck, vice president of hyperscale and high-performance computing at NVIDIA. "Vera gives AI agents the CPU performance to act in real time, executing code, processing data and coordinating complex tasks. SpaceXAI is taking this architecture from massive AI factories to the next frontier of computing in orbit."

Mike Nicolls, president of SpaceXAI, said Vera provides the CPU performance and memory bandwidth to run large amounts of orchestration, code and data processing while keeping GPUs focused on their core work. "That means higher-performance AI agents and more useful work from every watt of compute," he said.

The collaboration is intended to deliver one computing foundation across a wide range of environments, with Vera CPUs accelerating AI agents, Vera Rubin powering the infrastructure behind Grok and gigawatt-scale AI factories on Earth, and NVIDIA accelerated computing extending into orbital AI infrastructure.

NVIDIA noted that many of the products and features described remain in various stages and will be offered on a when-and-if-available basis. Development, release and timing of the technologies remain at NVIDIA's discretion, and no timeline was given for the Starmind satellite or the gigawatt-scale expansion.