AI News
Real Time

NVIDIA Unveils Space Computing Platform with Space-1 Vera Rubin Module to Power Orbital AI

📰 Press Release: NVIDIA Brings AI to Orbit with New Space Computing PlatformSAN JOSE, Calif. — GTC — March 16, 2026 — Nvidia today announced the...

📰 Press Release: NVIDIA Brings AI to Orbit with New Space Computing Platform

SAN JOSE, Calif. — GTC — March 16, 2026 — Nvidia today announced the dawn of a new era in space innovation with its latest generation of accelerated computing platforms. Designed to deliver data center-grade performance in Size, Weight, and Power (SWaP) constrained environments, these platforms will power orbital data centers (ODCs), advanced geospatial intelligence, and autonomous space Operations.
By enabling AI APPlications to run seamlessly from the ground to space—and between satellites—NVIDIA is helping to support increasingly complex mission scenarios. The new lineup includes the NVIDIA Space-1 Vera Rubin moduleIGX Thor, and Jetson Orin platforms.
"Space computing, the final frontier, is here. As we deploy satellite constellations and explore deeper into space, intelligence must exist wherever data is generated," said Jensen Huang, founder and CEO of NVIDIA. "AI processing across space and ground systems enables real-time sensing, decision-making, and autonomy, transforming Orbital Data Centers into tools for discovery and spacecraft into autonomous navigation systems. With our partners, we are extending NVIDIA beyond Earthboldly bringing Intelligence to places it has never been before."

🚀 Powering the Next Generation of Space Missions

Industry leaders including AetherflUX, Axiom Space, Kepler Communications, planet, Sophia Space, and Starcloud are already leverAGIng NVIDIA’s accelerated computing platforms to support next-generation space missions across orbital and ground environments.
  • Aetherflux: "Aetherflux is pioneering a new paradigm for space energy and computing. The NVIDIA Space-1 Vera Rubin module delivers high-performance, energy-efficient AI Compute in the orbital edge environment via solar power. This enables autonomous space operations and mission-critical services, laying the foundation for scalable space AI infrastructure beyond Earth," said Baiju Bhatt, Founder and CEO of Aetherflux.

  • Kepler Communications: "Kepler is building a next-generation space data network for real-time connectivity in orbit. NVIDIA Jetson ORIN provides our satellites with advanced AI capabilities to intelligently manage and route data across constellations. This transforms our network into a smarter, more efficient platform, reducing latency globally while providing SECure and reliable connectivity," said Mina Mitry, CEO of Kepler Communications.

  • Planet: "Planet images the Earth every day, and we need the world's most powerful compute to handle this data challenge. By integrating NVIDIA’s accelerated platforms from space to ground, we are drastically enhancing our ability to index the physical world. With the NVIDIA CorrDiff AI model, we can turn raw pixel data into ACTionable insights in near real-time. Together, we are achieving a revolutionary leap in planetary intelligence to help humanity make smarter decisions and keep pace with global changes," said Will Marshall, Co-founder and CEO of Planet.

  • Sophia Space: "Sophia Space is focused on building modular, passively cooled hosted compute platforms, giving customers dedicated infrastructure to run applications directly in space. NVIDIA Jetson Orin allows us to deeply embed AI capabilities into this infrastructure, supporting real-time processing and autonomy within strict SWaP constraints. This brings cloud-like flexibility to space and makes orbital computing commercially viable," said Rob DeMillo, CEO of Sophia Space.

  • Starcloud: "Starcloud is building custom Orbital data centers to deliver cloud and AI Infrastructure directly in space. With NVIDIA Technology, we can provide true hyperscale AI computing in orbit, processing data at the source to reduce reliance on downlinks. For the first time, this helps customers run training and inference workloads in space—a critical step in seamlessly extending global cloud services to orbit," said Philip Johnston, CEO of Starcloud.

🌌 AI-Enabled On-Orbit Infrastructure

The rapid growth of the commercial space industry has created a surging demand for real-time data processing in orbit.
  • NVIDIA Space-1 Vera Rubin Module: This module delivers massive Data Center-grade AI capabilities, enabling large language models (LLMs) and advanced foundation models to run directly in space. Its tightly integrated CPU-GPU architecture and high-bandwidth interconnects provide the necessary performance and mEMOry to process massive data streams from space instruments in real time. By introducing hyperscale AI capabilities to orbital platforms, the Space-1 Vera Rubin module enables on-orbit data analysis, autonomous scientific discovery, and rapid insight generation.

  • NVIDIA IGX Thor: Designed for next-generation mission-critical edge environments, this energy-efficient platform offers industrial-grade durability and enterprise-grade software support. It supports Real-time AI processing, functional safety, secure boot, and autonomous operations. This allows spacecraft to process sensor data locally, optimizing bandwidth usage and improving responsiveness while seamlessly complementing and extending ground control systems.

  • NVIDIA Jetson Orin: Built as an ultra-compact, energy-efficient module for edge deployment, Jetson Orin supports high-performance AI Inference. Optimized for SWaP-constrained environments, it enables real-time processing of vision, navigation, and sensor data directly on spacecraft to reduce latency and optimize bandwidth. The platform's AI software ecosystem and CUDA acceleration make Jetson Orin ideal for satellites, in-orbit service vehicles, and space sensing platforms that require intelligent responsive computing integrated with ground operations.

🌍 Advancing Geospatial Intelligence from the Ground

As the space ecosystem expands, so does the volume of data generated. While on-orbit computing enhances the real-time processing capabilities of geospatial sensing satellites (such as imaging sensors, radar, and RF sensors), the majority of collected data is still aggregated on Earth into hundreds of petabytes of Historical Archives to support large-scale geospatial trend analysis.
Traditionally, ground-based geospatial image processing systems have run on CPUs, resulting in long compute turnaround times. Compared to traditional architectures, NVIDIA RTX PRO Blackwell Server GPUs significantly accelerate ground processing speeds, delivering up to 100x performance improvement when analyzing massive image archives.
LeveRAGing the flexibility of CUDA, geospatial intelligence customers can adjust their processing power across the cloud, edge ground stations, and orbit. They can also rapidly integrate new AI capabilities to dynamically extract insights from massive image archives for:
  • Disaster Response & Environmental Monitoring: AI-accelerated processing of high-resolution images enables instant identification and rapid warning for wildfires, floods, and oil spills.

  • CLImate & Weather Forecasting: Flexible and precise tracking of weather patterns and long-term climate change, providing in-depth analysis of atmospheric data.

  • Infrastructure & Resource Management: Automating complex object detection and trend analysis to support autonomous monitoring of global energy grids, transportation networks, and agricultural ecosystem health.

★★★★★
★★★★★
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!