Meta's Space Llama AI Reaches the ISS National Laboratory - Know Details Here

Meta’s open-source AI model, Space Llama, is heading to space. Along with Booz Allen, a variant of Llama 3.2 will be deployed in the International Space Station (ISS) National Laboratory, where astronauts carry out vital experiments that can’t be done on Earth.

Llama was selected due to its flexibility and ability to run offline, as there is no internet access on the ISS. Being open source also guarantees it is more secure and lower cost while allowing researchers to easily adapt without connectivity to Earth .

Also read : After almost a year of waiting, Meta AI is rolling out in Europe

The project, called Space Llama, integrates Booz Allen’s A2E2™ system, HPE’s Spaceborne Computer-2, and NVIDIA’s accelerated computing technologies. This unique tech stack allows astronauts to process data, make decisions faster, and retrieve critical information without the need for live connectivity. 

Space Llama supports both generative and multimodal AI, capable of understanding text, images, and audio.

Space Llama builds on Booz Allen’s successful AI deployment in space in 2024 and marks a step forward in preparing for future lunar and Mars missions. It highlights how open source AI can drive American innovation, offering durable solutions in the harshest environments. 

This launch signals just the beginning for AI’s role in advancing space exploration.