At its highly anticipated GTC conference, NVIDIA introduced a groundbreaking suite of AI-driven technologies, including the Isaac GR00T N1—the world’s first open, fully customizable foundation model designed for humanoid robotics.
Alongside this launch, NVIDIA revealed the Isaac GR00T Blueprint for synthetic data generation and Newton, an advanced physics engine developed in collaboration with Google DeepMind and Disney Research.
The announcements mark a significant milestone in NVIDIA’s efforts to bridge AI with real-world applications, opening new possibilities across industries such as robotics, healthcare, and autonomous systems.
The Future of Humanoid Robotics
A highlight of the event, Isaac GR00T N1 is NVIDIA’s first fully customizable foundation model designed to enhance humanoid robotics capabilities.
This model, which will be pre-trained and made available to robotics developers worldwide, aims to address the growing labor shortages, estimated to exceed 50 million workers globally.
At its core, GR00T N1 features a dual-system architecture inspired by human cognition:
-
System 1 executes rapid, instinctive actions.
-
System 2 performs deliberate, analytical decision-making using a vision-language model to interpret environments and instructions.
This sophisticated approach allows the model to generalize tasks such as grasping, manipulating, and seamlessly transferring objects between its arms with remarkable precision.
During his keynote, NVIDIA CEO Jensen Huang demonstrated GR00T N1 in action, showcasing a humanoid robot from 1X Technologies autonomously performing household tasks using post-trained policies.
“The age of generalist robotics is here,” Huang stated. “With NVIDIA Isaac GR00T N1 and our new AI-driven frameworks, robotics developers will unlock new frontiers in AI-powered automation.”
Early adopters of GR00T N1 include 1X Technologies, Agility Robotics, Boston Dynamics, Mentee Robotics, and NEURA Robotics.
Newton: A Next-Generation Physics Engine for Robotics
In a major collaboration with Google DeepMind and Disney Research, NVIDIA introduced Newton, an open-source physics engine optimized for robot learning.
Built on the NVIDIA Warp framework, Newton integrates seamlessly with simulation tools such as Google DeepMind’s MuJoCo and NVIDIA Isaac Lab.
Disney Research plans to leverage Newton to enhance its robotic character platform, which powers expressive entertainment robots like Star Wars-inspired BDX droids. Kyle Laughlin, SVP of Walt Disney Imagineering R&D, highlighted the engine’s impact: “This collaboration will allow us to create robotic characters that are more expressive and engaging than ever before.”
Newton is set to launch later this year.
Cosmos World Foundation Models
NVIDIA also introduced Cosmos World Foundation Models (WFMs), designed to revolutionize AI-driven world generation and reasoning. A major breakthrough for robotics and autonomous systems, Cosmos introduces an open, customizable reasoning model, enabling AI to predict real-world interactions more accurately.
Key Components of Cosmos WFMs:
-
Cosmos Transfer: Transforms structured video inputs (e.g., segmentation maps and LiDAR scans) into realistic synthetic data, enhancing AI training accuracy.
-
Cosmos Predict: Enables AI to predict motion and actions from images, text, and video, improving decision-making for robots and self-driving cars.
-
Cosmos Reason: An open-access model allowing AI to understand and interact with dynamic environments.
Cosmos WFMs are now available via NVIDIA’s API catalog and Google Cloud’s Vertex AI Model Garden, with additional models accessible on Hugging Face and GitHub.
Isaac for Healthcare
Expanding its AI reach, NVIDIA partnered with GE HealthCare to drive innovation in autonomous medical imaging through Isaac for Healthcare.
This simulation platform includes pre-trained AI models and physics-based simulations of medical environments, aiming to address the global shortage of diagnostic imaging access.
Kimberly Powell, VP of Healthcare at NVIDIA, emphasized the platform’s importance: “The demand for healthcare services far exceeds supply. With Isaac for Healthcare, we’re enabling lifesaving medical devices to operate autonomously and improve access to healthcare worldwide.”
Built on NVIDIA DGX, Omniverse, and Holoscan, the platform is already being explored by early adopters such as Moon Surgical, Neptune Medical, and Xcath.
