🤖 Physical AI (Robotics & Embodied AI) 분야의 오픈소스 모델, 데이터셋, 시뮬레이터를 체계적으로 정리한 큐레이션 리스트. A curated list of open-source models, datasets, and simulators for Physical AI (Robotics & Embodied AI).
스타 수 기준 내림차순 정렬 | Sorted by GitHub stars (auto-updated weekly)
| Name | Organization | Year | Category | Hardware | Learning | ⭐ Stars | Links |
|---|---|---|---|---|---|---|---|
| LeRobot | Hugging Face | 2024 | manipulation | manipulator | IL, diffusion, VLA | 23,337 | 📄 🤗 |
| π0 (pi-zero) | Physical Intelligence | 2024 | manipulation, whole-body | manipulator, humanoid | VLA, diffusion | 11,365 | 📄 🤗 |
| GR00T N1 | NVIDIA | 2025 | manipulation, whole-body | humanoid | VLA, IL | 6,729 | 📄 🤗 |
| OpenVLA | Stanford / UC Berkeley | 2024 | manipulation | manipulator | VLA | 5,936 | 📄 🤗 |
| Mobile ALOHA | Stanford | 2024 | manipulation, navigation, whole-body | mobile, manipulator | IL | 4,407 | 📄 |
| Diffusion Policy | Columbia University | 2023 | manipulation | manipulator | IL, diffusion | 4,026 | 📄 |
| ACT (Action Chunking with Transformers) | Stanford | 2023 | manipulation, dexterous | manipulator | IL | 1,879 | 📄 |
| Octo | UC Berkeley / Stanford / CMU / others | 2023 | manipulation | manipulator, mobile | IL, VLA | 1,623 | 📄 🤗 |
| HumanPlus | Stanford | 2024 | manipulation, whole-body | humanoid | IL | 837 | 📄 |
| RoboFlamingo | ByteDance | 2023 | manipulation | manipulator | VLA | 429 | 📄 |
| GR-1 | BAAI / Beijing Academy of AI | 2024 | manipulation | manipulator | VLA, IL | 309 | 📄 |
| CrossFormer | UC Berkeley / others | 2024 | manipulation | manipulator, mobile | IL, VLA | 282 | 📄 🤗 |
스타 수 기준 내림차순 정렬 | Sorted by GitHub stars (auto-updated weekly)
| Name | Organization | Year | Category | Source | Modality | Trajectories | ⭐ Stars | Links |
|---|---|---|---|---|---|---|---|---|
| ManiSkill2 | UC San Diego / Shanghai AI Lab | 2023 | manipulation, dexterous | simulation | rgb, rgbd, proprioception | 36,000 | 2,797 | 📄 🤗 |
| Meta-World | Stanford / Berkeley | 2019 | manipulation | simulation | proprioception | — | 1,793 | 📄 |
| Open X-Embodiment (OXE) | Google DeepMind / RT-X Team | 2023 | manipulation | real, teleoperation | rgb, rgbd, proprioception | 1,000,000 | 1,782 | 📄 🤗 |
| LIBERO | UMass Amherst / Bosch | 2023 | manipulation | simulation | rgb, proprioception | 130,000 | 1,742 | 📄 🤗 |
| HumanoidBench | CMU / UC San Diego / MIT | 2024 | locomotion, manipulation, whole-body | simulation | rgb, proprioception | — | 748 | 📄 |
| RoboAgent Dataset | CMU | 2023 | manipulation | real, teleoperation | rgb, proprioception | 7,500 | 390 | 📄 |
| Language-Table | 2023 | manipulation | real, simulation, teleoperation | rgb, proprioception | 600,000 | 356 | 📄 🤗 | |
| DROID | UC Berkeley / Stanford / others | 2024 | manipulation | real, teleoperation | rgb, rgbd, proprioception | 76,000 | 345 | 📄 🤗 |
| BridgeData V2 | UC Berkeley | 2023 | manipulation | real, teleoperation | rgb, proprioception | 60,096 | 279 | 📄 🤗 |
| RH20T | Shanghai AI Lab | 2023 | manipulation, dexterous | real, teleoperation | rgb, rgbd, tactile, proprioception, audio | 110,000 | 105 | 📄 |
스타 수 기준 내림차순 정렬 | Sorted by GitHub stars (auto-updated weekly)
| Name | Organization | Year | Type | GPU | ROS2 | Language | ⭐ Stars | Paper |
|---|---|---|---|---|---|---|---|---|
| Genesis | Genesis Authors (MIT/CMU/Stanford) | 2024 | physics_engine | ✅ | — | python | 28,523 | 📄 |
| MuJoCo | DeepMind / Google | 2022 | physics_engine | — | — | c, python, c++ | 8,700 | 📄 |
| Isaac Lab | NVIDIA | 2023 | rl_framework | ✅ | ✅ | python | 6,962 | 📄 |
| dm_control | DeepMind | 2018 | benchmark | — | — | python | 3,800 | 📄 |
| Brax | 2021 | physics_engine | ✅ | — | python | 2,400 | 📄 | |
| Robosuite | Stanford / UT Austin | 2020 | benchmark | — | — | python | 1,700 | 📄 |
| RoboCasa | UT Austin | 2024 | full_stack | — | — | python | 1,400 | 📄 |
| SimplerEnv | UC Berkeley / Google | 2024 | benchmark | — | — | python | 950 | 📄 |
| SAPIEN | UC San Diego / others | 2020 | physics_engine | ✅ | — | python, c++ | 900 | 📄 |
새 항목을 추가하려면 GitHub Issue를 열어주세요. To add a new entry, please open a GitHub Issue:
이슈가 등록되면 봇이 자동으로 PR을 생성하고, 관리자가 검토 후 머지합니다. A bot will automatically create a PR from your issue for admin review.
| Field | Valid Values |
|---|---|
categories |
manipulation · locomotion · navigation · dexterous · whole-body · aerial |
hardware |
manipulator · humanoid · quadruped · biped · mobile · drone · hand |
learning |
VLA · IL · RL · diffusion · world_model · sim2real |
framework |
pytorch · jax · tensorflow |
communication |
ros2 · grpc · lcm · zenoh |
| Field | Valid Values |
|---|---|
source |
real · simulation · teleoperation · human_demo · mocap |
modality |
rgb · rgbd · depth · lidar · tactile · proprioception · audio · force_torque |
📊 Stats auto-updated every Sunday via GitHub Actions · README auto-generated by scripts/generate_site.py