SpaceXAI to Deploy Nvidia Vera CPUs and Orbital NVL72 for Grok AI

SpaceXAI is deploying standalone Nvidia Vera CPUs to power agentic workloads for the Grok AI, marking a strategic expansion of its compute capabilities. The company also plans to launch an optimized Vera Rubin NVL72 system into orbit aboard the Starmind satellite.

According to reports from Tom's Hardware and other industry outlets, the deployment of standalone Vera CPUs is specifically designed to handle "agentic workloads" for Grok. This hardware integration extends beyond Earth, with the planned deployment of the Vera Rubin NVL72 system on the Starmind satellite, signaling a push to bring high-performance AI infrastructure into space.

The Shift to Agentic AI

This development follows the rapid expansion of xAI's compute clusters and a deepening strategic partnership between Elon Musk's ventures and Nvidia. The specific focus on agentic workloads indicates a fundamental shift in the evolution of Grok. While previous iterations focused primarily on chat-based interfaces and information retrieval, agentic AI refers to systems capable of autonomous action and complex task execution without constant human prompting.

Implications for Orbital Computing

Moving high-end hardware like the NVL72 into space represents a significant leap in orbital computing. By processing data directly on satellites rather than beaming raw information back to terrestrial data centers, SpaceXAI can drastically reduce latency for space-based operations. This capability enables advanced AI inference and data processing to happen in real-time within the orbital environment, potentially transforming how satellite constellations manage telemetry and environmental data.

Nvidia's Expanding Footprint

For Nvidia, these deployments highlight the company's growing influence in specialized, non-terrestrial environments. While Nvidia dominates the terrestrial data center market, the integration of its Vera architecture into orbital hardware proves the versatility of its silicon in extreme conditions. This move positions Nvidia as the primary hardware provider for the emerging field of space-based AI infrastructure.

Future Outlook

Industry observers are now watching for the specific performance benchmarks of the Vera CPUs in agentic roles and the successful deployment timeline of the Starmind satellite. While the hardware plans are clear, the full scope of the autonomous capabilities Grok will gain from this specialized compute remains to be seen.