NASA just deployed Gemma to Space 🛰️
NASA+Loft Orbital ran Gemma 3 in orbit onboard a Loft Orbital's YAM-9 satellite. They use LangGraph for agent orchestration and Gemma 3 to generate short English summaries right from space
A Loft Orbital YAM-9 spacecraft hosted the first reported run of a vision-language model in orbit, where NASA JPL's NAVI-Orbital harness let Gemma 3 analyze Earth imagery onboard and flag areas of interest without shipping raw data to the ground.
NASA just deployed Gemma to Space 🛰️
NASA+Loft Orbital ran Gemma 3 in orbit onboard a Loft Orbital's YAM-9 satellite. They use LangGraph for agent orchestration and Gemma 3 to generate short English summaries right from space
By handling queries and classification locally on the Nvidia Jetson hardware, the setup points to lower downlink volumes and faster response for monitoring tasks, though exact bandwidth savings remain unreported.
LangGraph managed the autonomous workflow in orbit, but broader questions linger around how such systems scale when performance numbers like latency or accuracy stay undisclosed.
Users praise NASA running the Gemma 3 model on a Loft Orbital satellite in orbit as the best example of what small language models can achieve.
Read more about it in https://techcrunch.com/2026/06/15/a-satellite-just-learned-to-find-things-on-its-own-heres-what-that-means/
NASA just deployed Gemma to Space 🛰️
NASA+Loft Orbital ran Gemma 3 in orbit onboard a Loft Orbital's YAM-9 satellite. They use LangGraph for agent orchestration and Gemma 3 to generate short English summaries right from space
Gemma running in space! I don't think we had this on the radar when we trained the model!
NASA just deployed Gemma to Space 🛰️
NASA+Loft Orbital ran Gemma 3 in orbit onboard a Loft Orbital's YAM-9 satellite. They use LangGraph for agent orchestration and Gemma 3 to generate short English summaries right from space

@osanseviero This is the best example of what is possible with small language models!!

@osanseviero ⏫