LeRobot Goes to Driving School: World's Largest Open-Source Self-Driving Dataset
Hugging Face's LeRobot framework has been extended to include what is claimed to be the world's largest open-source self-driving dataset, released via a blog post on March 11, 2025. The dataset is intended to accelerate research in autonomous driving by providing large-scale, openly accessible driving data. This represents a significant expansion of LeRobot beyond its original robotics manipulation focus into the autonomous vehicle domain.
Related guides (3)
Related events (8)
LeRobot Community Datasets: The "ImageNet" of Robotics — When and How?
Hugging Face's LeRobot blog post discusses the vision and current state of building a large-scale community robotics dataset analogous to ImageNet for computer vision. The post examines what it would take to create a standardized, scalable dataset repository for robot learning, drawing on the LeRobot ecosystem. It addresses data collection formats, community contribution workflows, and the open challenges in making such a resource practically useful for training generalizable robot policies.
LeRobotDataset v3.0: Large-Scale Dataset Support for LeRobot
Hugging Face has released version 3.0 of the LeRobotDataset format, aimed at enabling large-scale robotics datasets within the LeRobot framework. The update introduces infrastructure improvements to support the storage, streaming, and management of significantly larger robot learning datasets. This is a tooling and data infrastructure milestone for the open-source robotics learning ecosystem built around LeRobot.
LeRobot v0.4.0: Supercharging OSS Robot Learning
Hugging Face released LeRobot v0.4.0, a major update to its open-source robot learning library. The release targets improvements in robotics policy training and deployment tooling within the open-source ecosystem. Specific capability changes and new features are not detailed in the provided body, but the version bump signals continued active development of the platform.
LeRobot v0.5.0: Scaling Every Dimension
Hugging Face released LeRobot v0.5.0, a major update to its open-source robotics learning library. The release focuses on scaling across multiple dimensions of the robotics ML pipeline. As a tier-2 source with no body content available, specific technical details of the update are not accessible from this item.
Scaling Robotics Datasets with Video Encoding
Hugging Face published a blog post on using video encoding techniques to scale robotics datasets. The post addresses the practical challenge of storing and transmitting large-scale robot learning data efficiently. Video compression is presented as a key infrastructure enabler for expanding robotics training corpora.
Ingredients for robotics research
OpenAI released eight simulated robotics environments and a Baselines implementation of Hindsight Experience Replay (HER), developed over the prior year for internal research. These environments were used to train models that transfer to physical robots. The release also included a set of research requests to guide community contributions in robotics.
Building a Healthcare Robot from Simulation to Deployment with NVIDIA Isaac
A Hugging Face blog post describes a project combining LeRobot and NVIDIA Isaac to develop a healthcare robot, covering the pipeline from simulation to real-world deployment. The post likely details how reinforcement learning or imitation learning techniques are applied in a medical robotics context. This represents a practical application of sim-to-real transfer methods in a high-stakes domain.
SmolVLA: Efficient Vision-Language-Action Model trained on Lerobot Community Data
Hugging Face introduces SmolVLA, a compact Vision-Language-Action model designed for robotics control, trained on community-contributed data from the LeRobot ecosystem. The model targets efficient deployment on resource-constrained hardware while maintaining competitive manipulation performance. This release represents a continuation of Hugging Face's strategy to democratize robotics AI through open community data pipelines.