{"slug": "rp1-an-open-source-full-stack-platform-for-agentic-humanoids", "title": "RP1: An Open Source Full-Stack Platform for Agentic Humanoids", "summary": "RoboParty released RP1, a fully open-source full-stack humanoid robot platform spanning mechanical CAD, electronics, firmware, control and learning, with supported configurations of 24, 28, 30 and 42 degrees of freedom. The platform is distributed across repositories including rp1_hardware, ROMOMO, RP-hand, RP1_firmware, roboparty_deploy, RP1_description and roboparty_train, with AMP, BeyondMimic, MimicLite, UFO, PBFM and IBFM listed as coming soon. RoboParty states the project has \"no black boxes, nothing withheld — full control handed back to the developer,\" and directs contributions to individual sub-repositories rather than the overview repository.", "body_md": "**English**   /   [简体中文](https://github.com/Roboparty/RP1/blob/main/README.zh-CN.md)\n\n  From mechanical design and electronics to firmware, control and learning.\n\n  Built to be understood. Open to be built upon.\n\n**RP1 is RoboParty's fully open-source humanoid platform: a shared foundation for building, understanding and extending a real robot.**\n\nThe platform brings mechanical and electrical design, joint modules, firmware, robot models, deployment and learning into one connected development stack. Explore a single subsystem, reproduce a build, or develop your own control and learning workflows.\n\n| Hardware you can inspect | Software you can extend | Intelligence you can explore | \n|---|---|---|\n| Mechanical CAD, schematics, PCB layouts, Gerber files, BOM and assembly documentation. | Linux BSP, communication, device interfaces, robot descriptions and deployment modules. | Locomotion training, motion imitation, tracking and behavior-model projects. | \n\n<sub>RP1, photographed in the real world.</sub>\n\n**No black boxes, nothing withheld — full control handed back to the developer**\n\nRelease contents, supported configurations and setup instructions are maintained in each repository.\n\nContributions belong in the relevant sub-repository. This overview repository does not accept pull requests; please open issues and pull requests in the project you are working on.\n\n| Repository | Description | \n|---|---|\n| [**rp1_hardware**](https://github.com/Roboparty/rp1_hardware) | Mechanical source files, electronics, testing and assembly SOPs. | \n| [**ROMOMO**](https://github.com/Roboparty/ROMOMO) | Joint modules, firmware and host tools. | \n| [**RP-hand**](https://github.com/Roboparty/rp_dexhand) | Dexterous hand wiring, CAN-FD communication and SDK integration. | \n| [**RP1_TEST**](https://github.com/Roboparty/rp1_test) | RP1 full-machine testing. | \n| [**robot_supplychain**](https://github.com/Roboparty/robot_supplychain) | Component selection, inspection, quality gates and acceptance. | \n| [**roboparty_joystick**](https://github.com/Roboparty/roboparty_joystick) | RoboParty controller and app. | \n\n## **Robot configurations**\n\nSee the description repository's current release for supported robot configurations.\n\nSupported configurations include 24, 28, 30 and 42 DoF variants.\n\n| Repository | Description | \n|---|---|\n| [**roboparty_deploy**](https://github.com/Roboparty/roboparty_deploy) | Runtime SDK for IMU, motor, dexhand, inference and camera modules. | \n| [**RP1_description**](https://github.com/Roboparty/RP1_description) | URDF, MJCF, 3D assets and robot configurations. | \n| [**roboparty_train**](https://github.com/Roboparty/roboparty_train) | Locomotion, motion imitation and behavior foundation models. | \n| [**RP1_firmware**](https://github.com/Roboparty/RP1_firmware) | Linux BSP, USB-CAN bridge, low-level SDK and packages. | \n\nComing soon.\n\n| Project | Description | \n|---|---|\n| [**AMP**](https://github.com/Roboparty/roboparty_train) | Learns motion priors from reference motions through adversarial training, supporting natural-looking locomotion and control. | \n| [**BeyondMimic**](https://github.com/Roboparty/roboparty_train) | Tracks reference motions to reproduce expressive whole-body movements, including dance and motion-imitation sequences. | \n| [**MimicLite**](https://github.com/Roboparty/MimicLite) | Focuses on reference-motion tracking, translating target movements into coordinated whole-body robot actions. | \n| [**UFO**](https://github.com/Roboparty/UFO) | Explores a shared behavior foundation for learning and controlling a range of humanoid movements. | \n| [**PBFM**](https://github.com/Roboparty/P-BFM) | Connects environmental perception with a behavior foundation model to study movement in relation to the surrounding scene. | \n| [**IBFM**](https://github.com/Roboparty/I-BFM-Code) | Focuses on whole-body interaction with the environment, exploring coordinated movements for contact-rich tasks. | \n| [**INTACT-JEPA**](https://github.com/Roboparty/INTACT-JEPA) | Learns an intent-to-action mapping with JEPA, translating visual goals into control actions without online search. | \n\n## **Related papers**\n\n- **[P-BFM: Perceptive Behavioral Foundation Models for Humanoid Control via Unsupervised Reinforcement Learning](https://arxiv.org/abs/2610.15833)** — Xue Wang, Yitang Li, Yiming Zhang, Siqin Wang, Junhan Sun, Zhihao Liu, Yaojie Shen, Jingbo Yuan, Hao Zhao, Qingdu Li.*ICLR 2027*\n- **[I-BFM: Reward-Conditioned Robust Humanoid Interaction via Unsupervised Reinforcement Learning](https://arxiv.org/abs/2610.06129)** — Ziqi Han, Yitang Li, Junhan Sun, Fanrong Dong, Yaojie Shen, Lei Ye, Zetong Jing, Yongqi Zhang, Yiming Zhang, Xue Wang, Hao Zhao.*ICRA*\n- **[PredActor: Predictive Action Diffusion for Steerable Onboard Humanoid Control](https://arxiv.org/abs/2609.24840)** — Lei Ye, Haibo Gao, Yitang Li, Peng Xu, Zetong Jing, Junhan Sun, Fanrong Dong, Ziqi Han, Xue Wang, Jianhua Sun, Cewu Lu, Hao Zhao, Liang Ding.*ICRA*\n- **[From Hitter to Player: Separating Motor Execution and Strategy in Competitive Humanoid Table Tennis](https://arxiv.org/abs/2610.14067)** — Fanrong Dong, Ziyi Sun, Ziqi Han, Junhan Sun, Lei Ye, Zetong Jing, Hao Zhao, Shaoxun Liu, Rongrong Wang.*ICRA*\n- **[INTACT: Isomorphic Intent-to-Action Learning for Search-Free World Models](https://arxiv.org/abs/2607.26056)** — Junhan Sun, Hao Zhao, Guofeng Zhang.*ICLR*\n- **[VIStream: Bringing Inertial Sensing to Streaming Geometric Foundation Models](https://github.com/CN-MingZheLiu/vistream)** — Mingzhe Liu.*ICRA*\n\nMeet the community, share your work, and connect with RoboParty.\n\n**RoboParty**\n\n  Open hardware. Open intelligence. Shared possibility.\n\n[GPL-3.0](https://github.com/Roboparty/RP1/blob/main/LICENSE)   ·   © [RoboParty](https://roboparty.com)   ·   [Back to top](#top)\n\n<sub>See each sub-repository for its applicable license.</sub>", "url": "https://wpnews.pro/news/rp1-an-open-source-full-stack-platform-for-agentic-humanoids", "canonical_source": "https://github.com/Roboparty/RP1", "published_at": "2026-10-16 08:05:39+00:00", "updated_at": "2026-10-16 08:40:06.051387+00:00", "lang": "en", "topics": ["robotics", "ai-research", "developer-tools", "ai-agents"], "entities": ["RoboParty", "RP1", "rp1_hardware", "ROMOMO", "RP-hand", "RP1_firmware", "roboparty_train", "BeyondMimic"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/rp1-an-open-source-full-stack-platform-for-agentic-humanoids", "markdown": "https://wpnews.pro/news/rp1-an-open-source-full-stack-platform-for-agentic-humanoids.md", "text": "https://wpnews.pro/news/rp1-an-open-source-full-stack-platform-for-agentic-humanoids.txt", "jsonld": "https://wpnews.pro/news/rp1-an-open-source-full-stack-platform-for-agentic-humanoids.jsonld"}}