NVIDIA and Hugging Face Bring New Models and Frameworks to LeRobot for the Open Robotics Community

NVIDIA and Hugging Face are collaborating to bring new models and frameworks to LeRobot, an open-source library for robotics, to provide developers with a more accessible and standardized path for end-to-end robot development. The integrations include the NVIDIA Isaac GR00T 1.7 model and the NVIDIA Isaac Teleop framework, with NVIDIA Cosmos 3 planned soon. This collaboration aims to drive innovation and collaboration across the open robotics community by providing shared models, data, and workflows for training and evaluating robots. The practical implication for engineers building AI systems is that they can now leverage these open-source resources to accelerate their robotics development.

⚡ Key Takeaways

• NVIDIA Isaac GR00T 1.7 is the first open and commercially viable robot foundation model, making it easier to post-train and deploy models through LeRobot workflows.
• NVIDIA Isaac Teleop is an open-source framework for robot data collection, helping developers capture high-quality human demonstrations from external devices using standardized, interoperable formats.
• The largest open-source physical AI dataset, downloaded over 15 million times, includes over 350,000 real and simulated trajectories and 57 million grasps to help developers kickstart their robotics workflows.
• NVIDIA Isaac Sim and Isaac Lab-based simulation frameworks support the full robotics development loop, including setting up environments, generating robot data, testing policies, and validating behaviors.
• NVIDIA Cosmos 3, a frontier world foundation model for physical AI, is planned to be integrated into LeRobot soon, enabling developers to generate and augment robotics data, simulate scenarios, and support policy development.

This collaboration between NVIDIA and Hugging Face has the potential to significantly accelerate the development of robotics applications by providing developers with access to shared models, data, and workflows. By leveraging these open-source resources, engineers can reduce the time and cost associated with developing and deploying robotics systems.

✅ Practical Steps

1. Utilize the NVIDIA Isaac GR00T 1.7 model and NVIDIA Isaac Teleop framework in LeRobot to streamline robot development workflows.
2. Leverage the largest open-source physical AI dataset to kickstart robotics workflows and accelerate development.
3. Explore the NVIDIA Isaac Sim and Isaac Lab-based simulation frameworks to support the full robotics development loop.

Want the full story? Read the original article.