LLeRobot Latest coverage

LeRobot@LeRobotHF· 08:10SignalEditorial score6565

LeRobotHF has released the Gen-HumanWholeBody dataset on Hugging Face, containing 600 hours of whole-body manipulation data with high precision (~3 cm MPJPE), 16,976 real-world scenarios, and 1,281 human skills. The dataset is open-source and aims to advance research in whole-body motion learning and human-to-robot transfer.

GreenWAM: A World Foundation Model for Robotics

A research paper introduces GreenWAM, a world foundation model for robotics that unifies world modeling and action control. The model is trained on a unified action space covering 20 robot types and demonstrates strong performance in trajectory prediction, depth estimation, and zero-shot task success across multiple benchmarks and real-world robot tasks. The model's architecture integrates world modeling with action generation through a shared transformer-based framework.

LeRobot SO-101 Tabletop Environment in Isaac Sim 6.2

LeRobot releases an Isaac Sim 6.2 environment for the SO-101 robotic arm, featuring a tabletop setup with wooden cubes and a green box. The environment supports the same policies as the MuJoCo version, with front and wrist cameras, joint position control, and a reward system for placing cubes in the box. It is available under the Apache-2.0 license.

LeRobot@LeRobotHF· 09:47SignalEditorial score6565

LeRobot now supports remote inference, allowing policy computation to run on a GPU server or cloud while the robot continues moving. Cameras, motors, and recording remain on the robot's local computer, with actions streamed back. This setup enables smooth motion execution with or without real-time communication, using Zenoh routers for secure cross-network operation.

LeRobot@LeRobotHF· 09:16SignalEditorial score7575

LeRobot now supports remote inference, enabling policy execution on GPU servers or cloud while the robot continues to move. Cameras, motors, and recording remain on the robot's local computer, with actions streamed back. This allows for smooth motion with or without real-time communication, and supports deployment across lab networks or the internet via Zenoh routers.

FineART: A Fine-Grained Annotated Robotic Trajectory Dataset for Bimanual Manipulation

FineART is a dataset of 40,543 teleoperated episodes with dense subtask annotations, collected on a bimanual robot. It includes 533,913 subtask labels and 151 tasks, with human-written annotations for training hierarchical policies and language models.

LingBot-VLA 2.0 Base Model for LeRobot

LingBot-VLA 2.0 (base) is a pre-trained policy for visual-language agents, adapted for LeRobot. It uses a Qwen3-VL-4B backbone with a sparse-MoE Qwen2 action expert, trained via flow matching. The model is available under Apache-2.0 license and supports fine-tuning with dual-query distillation.

LeRobot@LeRobotHF· 13:34SignalEditorial score8585

LeRobot's new streaming loader is designed for training on large-scale robot datasets, allowing efficient processing of 100 TB+ data without full local video copies. It reads episodes directly from Hugging Face or Buckets, prepares samples on-the-fly, and balances memory and throughput for varied training batches.

LeRobot@LeRobotHF· 16:04SignalEditorial score6565

The author explores the LeRobot community dataset, using embeddings and FiftyOne to analyze 497 episodes across 50 robot types. They identify near-duplicates, filter sessions, and reveal cross-embodiment patterns like cloth folding. A curated subset is available for further exploration.

LeRobot: Unitree G1 Hardware Modifications Dataset

This repository contains 3D-printable hardware modifications for the Unitree G1 robot, including grippers, head cameras, and backpack enclosures. It includes STL files and assembly instructions for physical customization without firmware or control code.

Hy-Embodied-0.5-VLA — LeRobot Model Conversion

LeRobot has converted the Hy-Embodied-0.5-VLA-RoboTwin model, achieving full forward parity with the original policy. The model supports 16D dual-arm pose/gripper actions and passes all validation checks, though it cannot handle more complex action spaces.

Hy-Embodied-0.5-VLA — LeRobot Model Conversion

LeRobot has released a converted version of the Hy-Embodied-0.5-VLA-UMI model, ensuring full tensor compatibility and validation. The model supports a 16D dual-arm embodiment but cannot express 12D mobile-manipulator actions, indicating limitations in the current representation.

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

NVIDIA and Hugging Face are releasing the NVIDIA Isaac GR00T 1.7 model and Isaac Teleop framework into LeRobot, an open-source robotics library, to provide developers with shared tools for training, evaluating, and deploying robot foundation models. NVIDIA Cosmos 3, a frontier world model for physical AI, is also set to be integrated soon.

LeRobot - Model Repository Updates

LeRobot introduces a Vision-Language-Action (VLA) policy called PI0Fast, which predicts continuous robot actions via autoregressive next-token prediction over FAST action tokens. The repository includes training and inference code, along with instructions for evaluation in simulation and real-world settings.

LeRobot: π₀.₅ VLA Model Release for Long-Horizon Tasks

LeRobot has released a Vision-Language-Action (VLA) model, π₀.₅, trained on robot demonstrations and multimodal data for long-horizon tasks. The model supports flow-matching for action prediction and is compatible with the original reference code. It is intended for fine-tuning on specific use cases and includes both inference and training examples.

LeRobot Releases π₀.₅ Vision-Language-Action Model for Physical Intelligence Tasks

LeRobot has released π₀.₅, a Vision-Language-Action (VLA) model with open-world generalization capabilities, trained on robot demonstrations and multimodal data. The model supports flow-matching for action prediction and is intended as a base model for fine-tuning. Code and documentation are provided for inference and training.

LeRobot π₀ Model Release: Vision-Language-Action Foundation Model for Robot Control

LeRobot introduces π₀, a Vision-Language-Action (VLA) foundation model that jointly reasons over vision, language, and actions to control robots. The model is designed as a base architecture for enabling open-world generalization and includes training and inference instructions for deployment.

LeRobot Releases π₀.₅ (Pi05) Model for Vision-Language-Action Tasks

LeRobot has released the π₀.₅ (Pi05) model, a Vision-Language-Action (VLA) model with open-world generalization capabilities. It is trained on robot demonstrations and multimodal data to execute long-horizon tasks in real-world environments. The model supports flow-matching for action prediction and is intended as a base for fine-tuning on specific applications. The implementation is compatible with the original reference code and includes instructions for inference and training.

LeRobot/pi0_base: Vision-Language-Action Foundation Model for Robot Control

LeRobot releases the π₀ model, a Vision-Language-Action (VLA) foundation model that jointly reasons over vision, language, and actions to control robots. It serves as the base architecture for π₀.₅'s open-world generalization. The model supports training, inference, and real-world evaluation with detailed code examples and setup instructions.

SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics

SmolVLA is a compact, efficient Vision-Language-Action (VLA) model designed for affordable robotics, trainable on a single GPU and deployable on consumer hardware. It matches the performance of larger VLAs through community-driven data, with a reference implementation and training instructions provided.