{"id":135271,"date":"2026-03-10T16:43:40","date_gmt":"2026-03-10T16:43:40","guid":{"rendered":"https:\/\/blogs.nvidia.com\/?p=90166"},"modified":"2026-03-10T16:43:40","modified_gmt":"2026-03-10T16:43:40","slug":"as-open-models-spark-ai-boom-nvidia-jetson-brings-it-to-life-at-the-edge-2","status":"publish","type":"post","link":"http:\/\/webdomino.net\/index.php\/nvidia\/as-open-models-spark-ai-boom-nvidia-jetson-brings-it-to-life-at-the-edge-2\/","title":{"rendered":"As Open Models Spark AI Boom, NVIDIA Jetson Brings It to Life at the Edge"},"content":{"rendered":"
<\/div>

The <\/span>Cat 306 CR<\/span><\/a> mini-excavator weighs just under eight tons and fits inside a standard shipping container. It\u2019s the machine a contractor rents when the job site is tight: a utility trench near a foundation, a basement dig in a dense neighborhood.<\/span><\/p>\n

The cab is roughly the size of a phone booth. The operator sits close to the controls, two joysticks, multiple functions per hand. It takes time to learn. It takes longer to speed up.<\/span><\/p>\n

At CES earlier this year, that <\/span>machine answered questions<\/span><\/a>.<\/span>
\n<\/span>
\n<\/span>In the demo, the Cat AI Assistant ran on <\/span>NVIDIA Jetson Thor<\/span><\/a>, an edge AI platform built for real\u2011time inference in industrial and robotic systems, <\/span>NVIDIA Nemotron<\/span><\/a> speech models are used for fast and accurate natural voice interactions. Qwen3 4B, served locally via vLLM, interprets requests and generates responses with low latency, no cloud link required.<\/span><\/p>\n

Beyond enterprise innovation, open models unlock new possibilities for developers to build and experiment freely. Running OpenClaw on NVIDIA Jetson enables developers to create private, always-on AI assistants at the edge \u2014 with zero application programming interface cost and full data privacy.<\/span><\/p>\n

All Jetson developer kits support OpenClaw, offering the flexibility to switch across open models from 2B to 30B. With a frontier-class AI assistant running locally, users can power morning briefings, automate daily tasks, perform code reviews and control smart home systems \u2014 all in real time.<\/span><\/p>\n

From the Cloud to the Edge<\/h2>\n

For most of their recent history, open models lived where it was easiest to support them.\u00a0<\/span><\/p>\n

They ran in data centers, backed by elastic compute and persistent networks. Cloud deployments carry costs in latency and ongoing compute spend that scale with every query.<\/span><\/p>\n

Physical systems optimize for something else. Low latency because machines interact with people and environments. Limited power because devices have hard limits. And consistent behavior because variability introduces risk.<\/span><\/p>\n

There\u2019s also a supply question. Memory shortages have driven up costs across the industry. Jetson brings compute and memory together in a system-on-module, accelerating customer hardware design and making sourcing and validation easier than with discrete component approaches.<\/span><\/p>\n

And as models have grown more efficient, developers have also started asking a different question. Not which model performs best in isolation, but where it makes sense to run.\u00a0<\/span><\/p>\n

More often, the answer is on the device, starting from Jetson Orin Nano 8GB for entry-level generative AI models.\u00a0\u00a0<\/span><\/p>\n

Building Autonomous Physical AI Systems at Scale<\/h2>\n

For <\/span>physical AI<\/span><\/a> systems, generative AI models are expanding what\u2019s possible.\u00a0<\/span><\/p>\n

Caterpillar’s in-cab Cat AI Assistant, which is in development, runs speech and language models locally alongside trusted machine context, supporting operator guidance and safety features.<\/span><\/p>\n

At CES, <\/span>Franka Robotics<\/span> showed what that looks like in robotics. The <\/span>company\u2019s FR3 Duo dual-arm system ran the NVIDIA GR00T N1.6 model<\/span><\/a> end-to-end onboard, perception to motion, no task scripting. The policy executes locally.<\/span><\/p>\n