Futures
Access hundreds of perpetual contracts
TradFi
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Launchpad
Be early to the next big token project
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
AI Enters the "Space Age": Nvidia's Jensen Huang Releases Space-1 Rubin Module, Inference Computing Power Surges 25x
IT Home, March 17 — Today (March 17), at the 2026 GTC Conference held in San Jose, California, NVIDIA announced the launch of the “Space Computing” platform, which includes the Space-1Vera Rubin module, IGX Thor, and Jetson Orin.
NVIDIA founder and CEO Jensen Huang emphasized, “Space computing is the final frontier that has now arrived. As satellite constellations are deployed and deep space exploration advances, intelligence must exist where data is generated.”
IT Home quotes a blog post stating that the new platform successfully brings data center-level performance into space environments with extreme size, weight, and power constraints (SWaP), accelerating spacecraft transformation into autonomous navigation systems capable of real-time perception and decision-making.
Regarding core space hardware, the focus of this release is the Space-1Vera Rubin module. Compared to the existing H100 GPU, the new module equipped with Rubin GPU achieves up to 25 times the performance leap in space AI inference, enabling the direct operation of large language models and foundational models in space.
To reduce latency in data transmission between space and ground, NVIDIA states that the IGX Thor and Jetson Orin platforms, with their extremely compact size and ultra-high efficiency, provide powerful image sensing and data acceleration capabilities for edge computing in orbit.
On the ground data processing side, massive amounts of space imagery also require rapid analysis. To this end, NVIDIA has launched the RTX PRO6000 Blackwell server-grade GPU. When analyzing large historical image archives, this GPU’s throughput is 100 times higher than traditional CPU-based batch processing systems.
Leveraging the flexibility of the CUDA architecture, customers can dynamically extract vast amounts of data (603138), providing near real-time precise decision support for disaster response (such as wildfire warnings), climate prediction, and global power grid monitoring.
The space computing ecosystem has already gained widespread industry recognition. Leading aerospace companies such as Aetherflux, Kepler Communications, Planet, and Starcloud are building the next-generation space networks using NVIDIA platforms.
In terms of availability, the IGX Thor, Jetson Orin platforms, and RTX PRO6000 GPU are now officially on the market, while the Space-1Vera Rubin module will be launched at a later date, fully opening the commercialization and popularization of space computing.
NVIDIA GTC 2026 Conference Special