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

NVIDIA and Hugging Face today announced a joint effort to deliver new models and frameworks to LeRobot, the open-source robotics platform hosted by Hugging Face. The goal is to make robotics development more accessible to developers worldwide by reducing the cost and complexity of building physical AI systems.

Robotics development has long been gated by expensive and fragmented resources — from large-scale datasets and robot foundation models to simulation environments, computing power, and validation tools. By bringing together NVIDIA's expertise in accelerated computing and physical AI with Hugging Face's strength in open-source model distribution, the partnership aims to address these bottlenecks.

The new models and frameworks will be integrated directly into the LeRobot platform, giving developers free access to validated robotics models and training pipelines. Developers can now build on proven foundations rather than starting from scratch.

For the broader AI community, this marks a significant push to apply the open-source model that succeeded in NLP and computer vision to the hardware-intensive field of robotics. Physical AI — AI systems that perceive, reason, and act in the real world — is widely seen as the next frontier, and accessible tooling is critical to its progress.

NVIDIA has been steadily increasing its open-source investments in robotics, from the Isaac Sim simulation platform to this LeRobot collaboration. Hugging Face continues to expand its role as an AI open-source hub, now moving into the robotics domain.

The key question going forward is whether LeRobot can attract enough community contributions to become a true "model zoo" for robotics, similar to what Hugging Face achieved in NLP and vision. Developer adoption and ecosystem growth will determine whether this initiative lives up to its potential.