NVIDIA Expands Isaac With Agile ONE SN Humanoid and Robot Learning Tools

Oct 21, 2026
NVIDIA expanded its Isaac robotics platform with the Agile ONE SN reference humanoid, Isaac OS, Isaac Lab 3.0, new datasets and robot training workflows.

NVIDIA expanded its Isaac robotics platform with a reference humanoid, datasets, a runtime and updated learning tools, the company announced at GTC Berlin. Agile Robots introduced Agile ONE SN, a humanoid research platform built on Isaac GR00T and Jetson Thor, with availability planned for early 2027.

Agile ONE SN stands 1.4 meters tall, weighs 40 kilograms and has 31 degrees of freedom. It includes two tactile five finger hands, cameras, lidar sensors and inertial measurement units. Its Jetson Thor computer provides 2,070 FP4 teraflops and 128GB of unified memory, while planned NVIDIA Halos integration will add safety functions.

NVIDIA also released Isaac OS, an open source runtime that connects AI models, sensors, controls and robot hardware. Isaac Lab 3.0 is now generally available with support for multiple GPUs and the Newton and PhysX physics engines.

The SimReady Central Library provides more than 2,000 assets for robotics and factory simulations. NVIDIA also released the Human 1M Dataset and three open source DGX Station workflows for refining GR00T 1.7 models, training manipulation skills and adapting Cosmos world models for synthetic data generation with ChatGPT Astra.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Robotics Brief, Industrial AI Weekly or Daily AI Brief.

Also, consider following us on social media:

AI & Robotics AI Brief AI Brief (X)
Free newsletter

Industrial AI Weekly

The latest advancements in smart manufacturing, predictive maintenance, AI-driven quality control, supply chain & more.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.