VOOZH about

URL: https://www.nvidia.com/en-in/edge-computing/space-computing/

⇱ Space Computing: On-Orbit AI & Accelerated Computing | NVIDIA

Space Computing

Boost AI applications, from Earth to space, with the NVIDIA accelerated computing platform.

Overview

The Final Frontier of Space Computing Has Arrived

Powering next-generation missions across ground-to-space environments requires new AI infrastructure. NVIDIA is collaborating with space industry leaders to deploy on-board accelerated computing in size, weight, and power-constrained environments across NVIDIA® Jetson Orin™, IGX Thor™, and the Space-1 Vera Rubin module. This enables orbital compute platforms to process and fuse sensor data, generate geospatial intelligence, and drive autonomous space operations with minimal downlink latency and costs. At edge ground stations, accelerated computing of satellite data processing—including compositing, orthorectification, and atmospheric compensation—delivers orders of magnitude speedup over legacy-based CPU systems.

NVIDIA Launches Space Computing to Boost AI Into Orbit

Axiom Space, Cowboy Space Corporation, Kepler Communications, Planet Labs, Sophia Space, and Starcloud are using NVIDIA accelerated computing platforms to power space missions across orbital and ground environments.

NVIDIA Announces the Space-1 Vera Rubin Module

At GTC, NVIDIA CEO Jensen Huang announced NVIDIA's collaboration with commercial space leaders to advance geospatial intelligence and autonomous space operations. These include natural disaster response, environmental monitoring, climate and weather prediction, and critical infrastructure and resource management.

Benefits

Unlock a New Era of Orbital Compute

Discover new levels of energy efficiency and performance with NVIDIA accelerated computing.

Faster On-Board AI Compute

The NVIDIA Vera Rubin Space-1 module delivers up to 25x more AI compute per GPU for space-based inferencing and orbital data centers.

High-Throughput Ground Processing

The NVIDIA RTX PRO™ 6000 Blackwell Server Edition GPU provides up to 100x faster performance versus legacy CPU-based batch systems for geospatial imagery analysis.

Enhanced Energy Efficiency

NVIDIA Jetson™ and IGX Thor offer energy-efficient, high-performance AI inference, image sensing, and accelerated data processing.

Deploy Frontier Models

Use frontier models in space for on-orbit analytics, autonomous scientific discovery, and rapid insight generation.

Products

Bring Edge AI to Orbital Data Centers

Discover data center-class performance with a full-stack accelerated computing platform, optimized for size, weight, and power (SWaP)-constrained environments.

Space-1 Vera Rubin Module

  • The NVIDIA Space-1 Vera Rubin module delivers data center-class AI compute at scale, enabling frontier and foundation models to operate directly in space.
  • Its tightly integrated CPU-GPU architecture and high-bandwidth interconnect provide the performance and memory needed to process massive data streams from space-based instruments in real time or form the processing backbone of orbital data centers.
  • By bringing hyperscale AI capability into orbital compute platforms, the Space-1 Vera Rubin module unlocks on-orbit analytics, autonomous scientific discovery, and rapid insight generation.

Jetson Orin

  • NVIDIA Jetson Orin delivers high-performance AI inference in an ultra-compact, energy-efficient module built for edge deployment.
  • Jetson Orin enables real-time processing of vision, navigation, and sensor data directly onboard spacecraft, reducing latency and optimizing bandwidth.
  • Jetson Orin combines the NVIDIA Jetson platform’s AI software ecosystem and NVIDIA CUDA® acceleration. This makes it ideal for satellites, on-orbit servicing vehicles, and space-based sensing platforms that require intelligent, responsive computing while remaining integrated with ground operations.

IGX Thor

  • NVIDIA IGX Thor provides industrial-grade durability and enterprise software support in a power-efficient platform designed for next-generation, mission-critical edge environments.
  • It supports real-time AI processing, functional safety, secure boot, and autonomous operation. This lets spacecraft process sensor data locally, optimize bandwidth use, and enhance responsiveness, all while seamlessly complementing and extending the capabilities of ground control systems.

RTX PRO 6000 Blackwell Server Edition (BSE)

  • The NVIDIA RTX PRO 6000 Blackwell Server Edition GPU delivers massive acceleration for on-ground processing over traditional architectures.
  • NVIDIA data center platforms deliver high-throughput, on-demand ground processing for geospatial intelligence, delivering up to 100x faster performance versus legacy CPU-based batch systems when analyzing massive imagery archives.

Resources

Understanding the Space Computing Opportunity

Explore key use cases and content from NVIDIA and key partners in the commercial space industry.

  1. Use Cases
  2. NVIDIA Blogs
  3. Customer Announcements
  4. Technical Sessions

Earth Orbit (EO) and Infrared (IR) Imagery

Process EO and IR imagery in orbit to deliver real-time wildfire, climate, and infrastructure insights instead of downlinking hundreds of terabytes of raw data to ground.

Radio Frequency (RF) and Synthetic Aperture Radar (SAR)

Accelerate RF signal processing and spectrum intelligence in orbit for resilient, low-latency communications and payload operations. Run AI-enhanced radar and SAR processing in space for rapid detection, tracking, and change monitoring to support critical mission decisions.

Autonomous Space Operations

Process real-time sensor data locally for mission-critical autonomous spacecraft control within strict size, weight and power constraints.

Planet Labs

Planet is collaborating with NVIDIA to build a GPU-native AI engine that accelerates satellite imagery into real-time planetary intelligence and insights.

Kepler Communications

Kepler is using NVIDIA Jetson Orin to power the first space-based, scalable cloud infrastructure, processing data in-orbit across its optical constellation.

Cowboy Space Corporation

Cowboy Space Corporation is bringing together NVIDIA Space-1 Vera Rubin module, IGX Thor, and Jetson Orin to orbit to power AI-driven geospatial intelligence and autonomous space operations.

Sophia Space

Sophia Space is using NVIDIA Jetson Orin to embed AI into modular orbital compute platforms, enabling real-time processing and autonomous operations directly in space.

Firefly Aerospace

Firefly Aerospace is integrating NVIDIA Jetson on its Elytra spacecraft to process lunar imagery on-orbit, turning raw Moon data into real-time insights for its Ocula imaging service.

Ecosystem

Explore Space Companies Building Orbital Compute With NVIDIA

Ruggedized NVIDIA edge compute for space computing and orbital data centers.

FAQs About the NVIDIA Space Computing Platform

Space computing is the deployment of accelerated AI infrastructure across orbital platforms and ground stations. NVIDIA's full-stack space computing platform brings data-center-class AI compute into SWaP-constrained environments, enabling satellites to process sensor data and generate geospatial intelligence in orbit—without relying on constant data downlinks to Earth.

Orbital data centers are satellites with GPU-class hardware that process AI workloads in space instead of sending raw data to Earth. NVIDIA enables them through the Space-1 Vera Rubin module, delivering up to 25x more AI compute per GPU. Partners include Axiom Space, Cowboy Space Corporation, Firefly Aerospace, Kepler Communications, Muon Space, Sophia Space, Starcloud, and Vast.

Orbital compute is GPU-accelerated processing aboard satellites in low Earth orbit. Space-based sensors generate more data than can be economically downlinked, so moving AI inference to orbit reduces latency from hours to minutes. NVIDIA Jetson Orin and IGX Thor are the leading hardware platforms for orbital compute today.

On-orbit AI runs inference aboard the satellite, transmitting only processed insights—not raw data—to ground stations, cutting costly downlink bandwidth. NVIDIA Jetson Orin combines AI inference and sensor fusion in a spacecraft-ready module. Planet Labs uses NVIDIA to convert satellite imagery into real-time intelligence in orbit.

NVIDIA's space computing platform includes four products: Space-1 Vera Rubin module for orbital data centers (25x more AI compute per GPU); Jetson Orin for onboard spacecraft AI; IGX Thor for mission-critical edge environments; and RTX PRO 6000 Blackwell Server Edition for ground-based satellite data processing (100x faster than CPU-based systems).

Stay Connected

Get the Latest News in Edge Computing

Stay up-to-date with NVIDIA announcements and partner news in edge AI, including space computing.

Aethero's NxA Edge Computing Module based on the NVIDIA Jetson AGX Thor or Orin - the NxA is available in a modular or multiple redundant or a kubernetes-enabled distributed computing deployment for space operations.

Aitech S-A2300 COTS AI supercomputer leverages NVIDIA Jetson Orin to create a system specifically designed for Low Earth Orbit (LEO) missions.

EDGX Sterna is a flight-proven edge computer, powered by NVIDIA Jetson Orin NX, for cubesats to microsats, with industry leading compute per watt-hour across 5-year LEO SSO missions.

The Eizo Condor™ Thor 3U VPX Series brings rugged single board computers, powered by NVIDIA Jetson Thor, designed for real-time ISR, uncrewed systems, and multi-sensor fusion.

The WOLF-14T5 is a rugged 3U VPX single-board computer powered by NVIDIA Jetson AGX Orin and ConnectX-7, designed to deliver high-performance AI processing and up to 100 GbE networking for demanding, secure embedded applications.

Get the Latest News in Edge Computing

Welcome back. Not you?
Welcome back. Not you?