[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-2297":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":23,"hasPages":23,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":26,"readmeContent":27,"aiSummary":28,"trendingCount":16,"starSnapshotCount":16,"syncStatus":29,"lastSyncTime":30,"discoverSource":31},2297,"lerobot","huggingface\u002Flerobot","huggingface","🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning","https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Flerobot",null,"Python",24902,4780,159,380,0,31,223,962,165,120,"Apache License 2.0",false,"main",[],"2026-06-12 04:00:14","\u003Cp align=\"center\">\n  \u003Cimg alt=\"LeRobot, Hugging Face Robotics Library\" src=\".\u002Fmedia\u002Freadme\u002Flerobot-logo-thumbnail.png\" width=\"100%\">\n\u003C\u002Fp>\n\n\u003Cdiv align=\"center\">\n\n[![Tests](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot\u002Factions\u002Fworkflows\u002Flatest_deps_tests.yml\u002Fbadge.svg?branch=main)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot\u002Factions\u002Fworkflows\u002Flatest_deps_tests.yml?query=branch%3Amain)\n[![Tests](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot\u002Factions\u002Fworkflows\u002Fdocker_publish.yml\u002Fbadge.svg?branch=main)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot\u002Factions\u002Fworkflows\u002Fdocker_publish.yml?query=branch%3Amain)\n[![Python versions](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fpyversions\u002Flerobot)](https:\u002F\u002Fwww.python.org\u002Fdownloads\u002F)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache%202.0-blue.svg)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot\u002Fblob\u002Fmain\u002FLICENSE)\n[![Status](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fstatus\u002Flerobot)](https:\u002F\u002Fpypi.org\u002Fproject\u002Flerobot\u002F)\n[![Version](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Flerobot)](https:\u002F\u002Fpypi.org\u002Fproject\u002Flerobot\u002F)\n[![Contributor Covenant](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FContributor%20Covenant-v2.1-ff69b4.svg)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot\u002Fblob\u002Fmain\u002FCODE_OF_CONDUCT.md)\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-Join_Us-5865F2?style=flat&logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002Fq8Dzzpym3f)\n\n\u003C\u002Fdiv>\n\n**LeRobot** aims to provide models, datasets, and tools for real-world robotics in PyTorch. The goal is to lower the barrier to entry so that everyone can contribute to and benefit from shared datasets and pretrained models.\n\n🤗 A hardware-agnostic, Python-native interface that standardizes control across diverse platforms, from low-cost arms (SO-100) to humanoids.\n\n🤗 A standardized, scalable LeRobotDataset format (Parquet + MP4 or images) hosted on the Hugging Face Hub, enabling efficient storage, streaming and visualization of massive robotic datasets.\n\n🤗 State-of-the-art policies that have been shown to transfer to the real-world ready for training and deployment.\n\n🤗 Comprehensive support for the open-source ecosystem to democratize physical AI.\n\n## Quick Start\n\nLeRobot can be installed directly from PyPI.\n\n```bash\npip install lerobot\nlerobot-info\n```\n\n> [!IMPORTANT]\n> For detailed installation guide, please see the [Installation Documentation](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Flerobot\u002Finstallation).\n\n## Robots & Control\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\".\u002Fmedia\u002Freadme\u002Frobots_control_video.webp\" width=\"640px\" alt=\"Reachy 2 Demo\">\n\u003C\u002Fdiv>\n\nLeRobot provides a unified `Robot` class interface that decouples control logic from hardware specifics. It supports a wide range of robots and teleoperation devices.\n\n```python\nfrom lerobot.robots.myrobot import MyRobot\n\n# Connect to a robot\nrobot = MyRobot(config=...)\nrobot.connect()\n\n# Read observation and send action\nobs = robot.get_observation()\naction = model.select_action(obs)\nrobot.send_action(action)\n```\n\n**Supported Hardware:** SO100, LeKiwi, Koch, HopeJR, OMX, EarthRover, Reachy2, Gamepads, Keyboards, Phones, OpenARM, Unitree G1.\n\nWhile these devices are natively integrated into the LeRobot codebase, the library is designed to be extensible. You can easily implement the Robot interface to utilize LeRobot's data collection, training, and visualization tools for your own custom robot.\n\nFor detailed hardware setup guides, see the [Hardware Documentation](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Flerobot\u002Fintegrate_hardware).\n\n## LeRobot Dataset\n\nTo solve the data fragmentation problem in robotics, we utilize the **LeRobotDataset** format.\n\n- **Structure:** Synchronized MP4 videos (or images) for vision and Parquet files for state\u002Faction data.\n- **HF Hub Integration:** Explore thousands of robotics datasets on the [Hugging Face Hub](https:\u002F\u002Fhuggingface.co\u002Flerobot).\n- **Tools:** Seamlessly delete episodes, split by indices\u002Ffractions, add\u002Fremove features, and merge multiple datasets.\n\n```python\nfrom lerobot.datasets.lerobot_dataset import LeRobotDataset\n\n# Load a dataset from the Hub\ndataset = LeRobotDataset(\"lerobot\u002Faloha_mobile_cabinet\")\n\n# Access data (automatically handles video decoding)\nepisode_index=0\nprint(f\"{dataset[episode_index]['action'].shape=}\\n\")\n```\n\nLearn more about it in the [LeRobotDataset Documentation](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Flerobot\u002Flerobot-dataset-v3)\n\n## SoTA Models\n\nLeRobot implements state-of-the-art policies in pure PyTorch, covering Imitation Learning, Reinforcement Learning, and Vision-Language-Action (VLA) models, with more coming soon. It also provides you with the tools to instrument and inspect your training process.\n\n\u003Cp align=\"center\">\n  \u003Cimg alt=\"Gr00t Architecture\" src=\".\u002Fmedia\u002Freadme\u002FVLA_architecture.jpg\" width=\"640px\">\n\u003C\u002Fp>\n\nTraining a policy is as simple as running a script configuration:\n\n```bash\nlerobot-train \\\n  --policy=act \\\n  --dataset.repo_id=lerobot\u002Faloha_mobile_cabinet\n```\n\n| Category                   | Models                                                                                                                                                                                                                  |\n| -------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| **Imitation Learning**     | [ACT](.\u002Fdocs\u002Fsource\u002Fpolicy_act_README.md), [Diffusion](.\u002Fdocs\u002Fsource\u002Fpolicy_diffusion_README.md), [VQ-BeT](.\u002Fdocs\u002Fsource\u002Fpolicy_vqbet_README.md), [Multitask DiT Policy](.\u002Fdocs\u002Fsource\u002Fpolicy_multi_task_dit_README.md) |\n| **Reinforcement Learning** | [HIL-SERL](.\u002Fdocs\u002Fsource\u002Fhilserl.mdx), [TDMPC](.\u002Fdocs\u002Fsource\u002Fpolicy_tdmpc_README.md) & QC-FQL (coming soon)                                                                                                             |\n| **VLAs Models**            | [Pi0Fast](.\u002Fdocs\u002Fsource\u002Fpi0fast.mdx), [Pi0.5](.\u002Fdocs\u002Fsource\u002Fpi05.mdx), [GR00T N1.5](.\u002Fdocs\u002Fsource\u002Fpolicy_groot_README.md), [SmolVLA](.\u002Fdocs\u002Fsource\u002Fpolicy_smolvla_README.md), [XVLA](.\u002Fdocs\u002Fsource\u002Fxvla.mdx)            |\n\nSimilarly to the hardware, you can easily implement your own policy & leverage LeRobot's data collection, training, and visualization tools, and share your model to the HF Hub\n\nFor detailed policy setup guides, see the [Policy Documentation](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Flerobot\u002Fbring_your_own_policies).\n\n## Inference & Evaluation\n\nEvaluate your policies in simulation or on real hardware using the unified evaluation script. LeRobot supports standard benchmarks like **LIBERO**, **MetaWorld** and more to come.\n\n```bash\n# Evaluate a policy on the LIBERO benchmark\nlerobot-eval \\\n  --policy.path=lerobot\u002Fpi0_libero_finetuned \\\n  --env.type=libero \\\n  --env.task=libero_object \\\n  --eval.n_episodes=10\n```\n\nLearn how to implement your own simulation environment or benchmark and distribute it from the HF Hub by following the [EnvHub Documentation](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Flerobot\u002Fenvhub)\n\n## Resources\n\n- **[Documentation](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Flerobot\u002Findex):** The complete guide to tutorials & API.\n- **[Chinese Tutorials: LeRobot+SO-ARM101中文教程-同济子豪兄](https:\u002F\u002Fzihao-ai.feishu.cn\u002Fwiki\u002Fspace\u002F7589642043471924447)** Detailed doc for assembling, teleoperate, dataset, train, deploy. Verified by Seed Studio and 5 global hackathon players.\n- **[Discord](https:\u002F\u002Fdiscord.gg\u002Fq8Dzzpym3f):** Join the `LeRobot` server to discuss with the community.\n- **[X](https:\u002F\u002Fx.com\u002FLeRobotHF):** Follow us on X to stay up-to-date with the latest developments.\n- **[Robot Learning Tutorial](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Flerobot\u002Frobot-learning-tutorial):** A free, hands-on course to learn robot learning using LeRobot.\n\n## Citation\n\nIf you use LeRobot in your project, please cite the GitHub repository to acknowledge the ongoing development and contributors:\n\n```bibtex\n@misc{cadene2024lerobot,\n    author = {Cadene, Remi and Alibert, Simon and Soare, Alexander and Gallouedec, Quentin and Zouitine, Adil and Palma, Steven and Kooijmans, Pepijn and Aractingi, Michel and Shukor, Mustafa and Aubakirova, Dana and Russi, Martino and Capuano, Francesco and Pascal, Caroline and Choghari, Jade and Moss, Jess and Wolf, Thomas},\n    title = {LeRobot: State-of-the-art Machine Learning for Real-World Robotics in Pytorch},\n    howpublished = \"\\url{https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot}\",\n    year = {2024}\n}\n```\n\nIf you are referencing our research or the academic paper, please also cite our ICLR publication:\n\n\u003Cdetails>\n\u003Csummary>\u003Cb>ICLR 2026 Paper\u003C\u002Fb>\u003C\u002Fsummary>\n\n```bibtex\n@inproceedings{cadenelerobot,\n  title={LeRobot: An Open-Source Library for End-to-End Robot Learning},\n  author={Cadene, Remi and Alibert, Simon and Capuano, Francesco and Aractingi, Michel and Zouitine, Adil and Kooijmans, Pepijn and Choghari, Jade and Russi, Martino and Pascal, Caroline and Palma, Steven and Shukor, Mustafa and Moss, Jess and Soare, Alexander and Aubakirova, Dana and Lhoest, Quentin and Gallou\\'edec, Quentin and Wolf, Thomas},\n  booktitle={The Fourteenth International Conference on Learning Representations},\n  year={2026},\n  url={https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.22818}\n}\n```\n\n\u003C\u002Fdetails>\n\n## Contribute\n\nWe welcome contributions from everyone in the community! To get started, please read our [CONTRIBUTING.md](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot\u002Fblob\u002Fmain\u002FCONTRIBUTING.md) guide. Whether you're adding a new feature, improving documentation, or fixing a bug, your help and feedback are invaluable. We're incredibly excited about the future of open-source robotics and can't wait to work with you on what's next—thank you for your support!\n\n\u003Cp align=\"center\">\n  \u003Cimg alt=\"SO101 Video\" src=\".\u002Fmedia\u002Freadme\u002Fso100_video.webp\" width=\"640px\">\n\u003C\u002Fp>\n\n\u003Cdiv align=\"center\">\n\u003Csub>Built by the \u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002Flerobot\">LeRobot\u003C\u002Fa> team at \u003Ca href=\"https:\u002F\u002Fhuggingface.co\">Hugging Face\u003C\u002Fa> with ❤️\u003C\u002Fsub>\n\u003C\u002Fdiv>\n","LeRobot 是一个旨在通过端到端学习使机器人领域的AI更加易用的项目。它提供了模型、数据集和工具，支持从低成本机械臂到人形机器人的多种平台，并且具有硬件无关性，使用Python原生接口实现跨平台标准化控制。LeRobot采用Parquet + MP4或图像格式的标准数据集格式，便于大规模存储、流媒体传输及可视化处理，同时提供经过验证能够迁移到实际环境中的先进策略。适用于希望降低进入门槛、促进开放共享的研究人员与开发者，以及任何对物理AI感兴趣的个人或团队。",2,"2026-06-11 02:49:20","top_language"]