Meta, in partnership with Booz Allen Hamilton, has launched a customised version of its large language model, Llama 3.2, to the International Space Station (ISS) National Laboratory, a major leap toward bringing advanced AI assistance to astronauts.
What’s Happening?
Meta and Booz Allen Hamilton have deployed “Space Llama,” a pioneering AI tech stack, aboard the ISS.
The system runs locally on Hewlett Packard Enterprise’s Spaceborne Computer-2, accelerated by NVIDIA GPUs, and is integrated with Booz Allen’s modular A2E2 edge computing platform.
Space Llama enables astronauts to interact with technical manuals, station guidelines, and scientific documentation through text, audio, and images, all without relying on intermittent or bandwidth-limited ground communications.
Why Is This a Big Deal?
Autonomous AI Support: Space Llama allows astronauts to access and analyze essential information, troubleshoot onboard anomalies, and perform maintenance tasks more efficiently, reducing dependence on ground control and paper documentation.
Multimodal Capabilities: The model processes text, audio, and images in real time, enabling seamless, context-aware assistance for complex scientific and operational challenges.
Edge Computing in Space: By running entirely on local hardware, Space Llama overcomes the limitations of unreliable or delayed internet connectivity, a critical issue faced by the ISS and future deep-space missions.
Cost and Efficiency: The system is designed to be compact and energy-efficient, lowering operational costs and freeing up astronauts for mission-critical work.
A Glimpse Into the Future
Space Llama is seen as one of the first steps for future lunar and Mars missions, where autonomous, on-site AI will be essential. The project demonstrates how open-source AI can drive innovation and resilience in extreme environments, not just in space but also in industries facing similar connectivity and operational challenges on Earth.
“This is just the beginning. We see a future where open source AI models like Llama will play a vital role in fueling space exploration and research-enabling astronauts to undertake the next level of experimentation to solve complex scientific problems and drive innovation in entirely new ways.”
Ahmad Al-Dahle, Meta VP and Head of GenAI
Is It Time for Real-Life TARS?
While Space Llama is not a physical robot like TARS from Interstellar, it represents a significant step toward that vision. By providing astronauts with an always-available, multimodal AI assistant capable of understanding and responding to complex tasks in real time, the groundwork is being laid for more advanced, autonomous AI companions in space
AI is making a significant impact in critical areas of research and scientific advancement. Open-source AI projects like Meta LLaMA are opening doors to the widespread use of cutting-edge AI systems, helping to address long-standing challenges that have hindered progress across various scientific and professional fields.
Discover how AI can be applied to your area of interest or profession by filling out the enquiry form on our home page. Visit our website to learn more about our easy-to-learn courses in AI, Machine Learning, and much more.