# huggingface/lerobot

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/huggingface-lerobot).**

21,687 stars · 3,802 forks · Python · apache-2.0

## Links

- GitHub: https://github.com/huggingface/lerobot
- Homepage: https://huggingface.co/docs/lerobot
- awesome-repositories: https://awesome-repositories.com/repository/huggingface-lerobot.md

## Description

This project is a comprehensive research platform designed for the end-to-end lifecycle of robotic learning. It provides a modular framework for training neural network policies—specifically through imitation and reinforcement learning—and deploying them onto physical robotic hardware. By offering a unified interface for hardware abstraction, the platform decouples high-level control logic from the specific sensors and actuators of diverse robotic systems.

The framework distinguishes itself through a standardized approach to data and policy management. It utilizes a consistent schema for recording and sharing interaction data, which includes synchronized video and state information. To support complex training requirements, it features distributed optimization across multiple graphics processing units and a kinematic engine that handles coordinate transformations between joint space and Cartesian systems. These capabilities are complemented by a flexible architecture that allows for the modular design of vision-language-action models.

Beyond core training, the platform includes extensive utilities for data processing, such as observation standardization and action normalization, ensuring compatibility across different environments and hardware configurations. It also provides integrated tools for benchmarking performance through standardized rollout loops and evaluation scripts. For resource-constrained hardware, the system supports remote inference streaming, allowing computational workloads to be offloaded to external servers while maintaining real-time control.

## Tags

### Artificial Intelligence & ML

- [Expert Imitation Learning](https://awesome-repositories.com/f/artificial-intelligence-ml/expert-imitation-learning.md) — Trains transformer-based models on demonstration datasets to enable autonomous robotic manipulation.
- [Edge AI Model Deployment](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-deployment-and-serving/local-and-on-device-inference/edge-ai-model-deployment.md) — Executes optimized machine learning models on physical robotic hardware for autonomous tasks. ([source](https://huggingface.co/docs/lerobot))
- [Machine Learning Training](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/machine-learning-training.md) — Provides standardized pipelines for training machine learning models for robotic manipulation tasks.
- [Interaction Data Collectors](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/machine-learning-datasets/interaction-data-collectors.md) — Records sensor inputs and motor commands into standardized formats for training datasets. ([source](https://huggingface.co/docs/lerobot/index))
- [Robotic Policy Evaluators](https://awesome-repositories.com/f/artificial-intelligence-ml/performance-evaluation-tools/robotic-policy-evaluators.md) — Executes standardized benchmarks to test trained policies in simulated and physical environments. ([source](https://cdn.jsdelivr.net/gh/huggingface/lerobot@main/README.md))
- [Benchmarking Suites](https://awesome-repositories.com/f/artificial-intelligence-ml/benchmarking-suites.md) — Runs standardized rollout loops across tasks to aggregate performance metrics like success rates. ([source](https://huggingface.co/docs/lerobot/adding_benchmarks))
- [Distributed Training](https://awesome-repositories.com/f/artificial-intelligence-ml/distributed-training.md) — Distributes training workloads across multiple GPUs to accelerate robotic policy learning. ([source](https://huggingface.co/docs/lerobot/multi_gpu_training))
- [Distributed Training Orchestrators](https://awesome-repositories.com/f/artificial-intelligence-ml/distributed-training-orchestrators.md) — Parallelizes the optimization of complex robotic policies across multiple graphics processing units.
- [Inference Rollout Evaluation](https://awesome-repositories.com/f/artificial-intelligence-ml/inference-execution/inference-rollout-evaluation.md) — Runs trained models on physical hardware to perform inference and record episodes for performance assessment. ([source](https://huggingface.co/docs/lerobot/act))
- [Dataset Management Tools](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/dataset-management/dataset-management-tools.md) — Standardizes data storage using synchronized video and state files for efficient dataset management. ([source](https://cdn.jsdelivr.net/gh/huggingface/lerobot@main/README.md))
- [Modular Policy Architectures](https://awesome-repositories.com/f/artificial-intelligence-ml/modular-training-architectures/modular-policy-architectures.md) — Implements imitation, reinforcement, and vision-language-action models in a modular structure. ([source](https://cdn.jsdelivr.net/gh/huggingface/lerobot@main/README.md))
- [Policy Architectures](https://awesome-repositories.com/f/artificial-intelligence-ml/modular-training-architectures/policy-architectures.md) — Organizes neural network components into interchangeable blocks for varied learning models.
- [Action Coordinate Transformers](https://awesome-repositories.com/f/artificial-intelligence-ml/agentic-systems-frameworks/integration-deployment/agent-frameworks/agent-runtimes/agent-action-representations/action-coordinate-transformers.md) — Converts action data between coordinate systems to support different training requirements. ([source](https://huggingface.co/docs/lerobot/action_representations))
- [Observation Processors](https://awesome-repositories.com/f/artificial-intelligence-ml/observation-processing/observation-processors.md) — Applies environment-specific transformations to observation data to handle unique sensor formats. ([source](https://huggingface.co/docs/lerobot/adding_benchmarks))
- [Normalization Migrators](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/frameworks/model-construction/neural-network-layers/normalization-layers/normalization-migrators.md) — Extracts normalization layers from trained model weights into external processor pipelines. ([source](https://huggingface.co/docs/lerobot/backwardcomp))

### Hardware & IoT

- [Robot Learning Platforms](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems/robotics-drones/robotic-tooling/robot-learning-platforms.md) — Provides a comprehensive research platform for the end-to-end lifecycle of robotic learning.
- [Robotics Middleware](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems/robotics-middleware.md) — Provides a unified interface to decouple high-level control logic from specific robotic hardware implementations.
- [Real-time Policy Execution](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/hardware-in-the-loop-simulators/real-time-policy-execution.md) — Executes trained machine learning models on physical robots by mapping outputs to motor control signals in real time. ([source](https://huggingface.co/docs/lerobot/index))
- [Robotics And Autonomous Systems](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems.md) — Maps trained machine learning model outputs to motor control signals for real-time autonomous robotic behavior.
- [Kinematics](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems/motion-planning-control/kinematics.md) — Calculates forward and inverse kinematics to translate between joint space and Cartesian coordinate systems. ([source](https://huggingface.co/docs/lerobot/action_representations))
- [Robotics and Control](https://awesome-repositories.com/f/hardware-iot/embedded-robotics/robotics-autonomous-systems/robotics-drones/robotics-and-control.md) — Optimizes neural network models using interaction data to enable autonomous behavior on physical robots. ([source](https://huggingface.co/docs/lerobot/index))

### Operating Systems & Systems Programming

- [Hardware Abstraction Layers](https://awesome-repositories.com/f/operating-systems-systems-programming/hardware-interfacing-drivers/hardware-abstraction-layers.md) — Decouples high-level control logic from specific communication protocols of diverse robotic hardware. ([source](https://cdn.jsdelivr.net/gh/huggingface/lerobot@main/README.md))

### Software Engineering & Architecture

- [Robotics Libraries](https://awesome-repositories.com/f/software-engineering-architecture/application-frameworks/robotics-libraries.md) — Provides a modular library for training imitation and reinforcement learning models for physical robots.
- [Performance Benchmarking](https://awesome-repositories.com/f/software-engineering-architecture/performance-reliability/performance-engineering/performance-benchmarking.md) — Runs standardized evaluation loops and rollout tests to measure robotic policy performance.
- [Robotic Data Processors](https://awesome-repositories.com/f/software-engineering-architecture/data-normalization-layers/robotic-data-processors.md) — Provides external processors to decouple input and output scaling from core model weights.

### Networking & Communication

- [Remote Inference Streaming](https://awesome-repositories.com/f/networking-communication/real-time-telemetry-streams/remote-inference-streaming.md) — Offloads computational workloads to external servers while maintaining real-time control.
- [Remote Inference Streaming](https://awesome-repositories.com/f/networking-communication/remote-procedure-execution/remote-inference-streaming.md) — Streams robot observations to remote servers for real-time policy inference on resource-constrained hardware. ([source](https://huggingface.co/docs/lerobot/async))

### Data & Databases

- [Action Normalizers](https://awesome-repositories.com/f/data-databases/data-normalization/action-normalizers.md) — Scales raw action values into a normalized range for model training and inference. ([source](https://huggingface.co/docs/lerobot/action_representations))
- [Robotic Interaction Schemas](https://awesome-repositories.com/f/data-databases/structured-data-schemas/robotic-interaction-schemas.md) — Structures interaction data into synchronized video and state files for cross-environment compatibility.
- [Observation Standardizers](https://awesome-repositories.com/f/data-databases/data-governance-modeling/data-standardization/observation-standardizers.md) — Maps simulator-specific outputs to a unified key convention for policy compatibility. ([source](https://huggingface.co/docs/lerobot/adding_benchmarks))
- [Legacy Dataset Replayers](https://awesome-repositories.com/f/data-databases/dataset-transformations/legacy-dataset-replayers.md) — Applies coordinate transformations to historical data to ensure compatibility with updated hardware. ([source](https://huggingface.co/docs/lerobot/backwardcomp))

### Development Tools & Productivity

- [Simulator Wrappers](https://awesome-repositories.com/f/development-tools-productivity/third-party-integrations/simulator-wrappers.md) — Wraps third-party simulators into a standardized interface for robotic policy training and evaluation. ([source](https://huggingface.co/docs/lerobot/adding_benchmarks))