Nvidia announced its Vera Rubin Space-1 Module at GTC 2026, bringing AI computing to orbit for the first time. The module delivers up to 25x more AI compute than the H100 GPU for space-based inferencing, enabling orbital data centers, geospatial intelligence processing, and autonomous space operations.

"Space computing, the final frontier, has arrived," CEO Jensen Huang declared during Monday's keynote. The Space-1 Module combines two Rubin GPUs with a single Vera CPU, engineered for the size, weight, and power constraints of space. Cooling posed a unique challenge — without convection in the vacuum of space, the system relies entirely on radiation cooling.

Partners including Axiom Space, Planet Labs, Starcloud, and Aetherflux will use the platform for next-generation space missions. The module taps into space's virtually unlimited solar power to run AI workloads that would strain terrestrial infrastructure.