{"id":34624920,"url":"https://github.com/horizonrobotics/holomotion","last_synced_at":"2026-06-09T11:00:29.791Z","repository":{"id":307432043,"uuid":"1028843437","full_name":"HorizonRobotics/HoloMotion","owner":"HorizonRobotics","description":"HoloMotion: A Foundation Model for Whole-Body Humanoid Control","archived":false,"fork":false,"pushed_at":"2026-05-16T04:23:57.000Z","size":5082,"stargazers_count":445,"open_issues_count":0,"forks_count":30,"subscribers_count":6,"default_branch":"master","last_synced_at":"2026-05-16T06:39:37.525Z","etag":null,"topics":["foundation-models","humanoid-robot","motion-tracking","robotics","whole-body-control"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/HorizonRobotics.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":"NOTICE","maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-07-30T06:31:25.000Z","updated_at":"2026-05-16T04:24:01.000Z","dependencies_parsed_at":"2025-09-09T14:20:57.913Z","dependency_job_id":"848caff4-668b-4146-bf81-275dde617f3c","html_url":"https://github.com/HorizonRobotics/HoloMotion","commit_stats":null,"previous_names":["horizonrobotics/holomotion"],"tags_count":10,"template":false,"template_full_name":null,"purl":"pkg:github/HorizonRobotics/HoloMotion","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HorizonRobotics%2FHoloMotion","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HorizonRobotics%2FHoloMotion/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HorizonRobotics%2FHoloMotion/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HorizonRobotics%2FHoloMotion/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/HorizonRobotics","download_url":"https://codeload.github.com/HorizonRobotics/HoloMotion/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HorizonRobotics%2FHoloMotion/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34103357,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-09T02:00:06.510Z","response_time":63,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["foundation-models","humanoid-robot","motion-tracking","robotics","whole-body-control"],"created_at":"2025-12-24T15:52:27.819Z","updated_at":"2026-06-09T11:00:29.782Z","avatar_url":"https://github.com/HorizonRobotics.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\n\u003cimg src=\"assets/media/holomotion_logo_text.jpg\" alt=\"HoloMotion Logo\" width=\"600\"/\u003e\n\n[![Safari](https://img.shields.io/badge/Website-006CFF?logo=safari\u0026logoColor=fff)](https://horizonrobotics.github.io/robot_lab/holomotion/)\n[![HuggingFace](https://img.shields.io/badge/-HuggingFace-3B4252?style=flat\u0026logo=huggingface\u0026logoColor=)](https://huggingface.co/collections/HorizonRobotics/holomotion)\n[![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/HorizonRobotics/HoloMotion)\n[![WeChat](https://img.shields.io/badge/Wechat-7BB32E?logo=wechat\u0026logoColor=white)](https://horizonrobotics.feishu.cn/docx/Xs3cdEI8bo1EZuxUfzjckTgKn2c)\n[![arXiv](https://img.shields.io/badge/arXiv-2605.15336-red?logo=arxiv\u0026logoColor=white)](https://arxiv.org/abs/2605.15336)\n\n\n\u003c/div\u003e\n\n# News\n- [2026.05.15] HoloMotion v1.3 scales from 60M to 0.4B parameters and 80 to 2000+ hours of motion data, while improving policy inference from ~100 to ~300 FPS.\n- [2026.04.04] HoloMotion v1.2 provides pre-trained motion tracking and velocity tracking models for the community to deploy directly.\n\n\n# Why HoloMotion?\n\n## Larger, Faster and Stronger\n\nHoloMotion scales humanoid whole-body control through a reference-conditioned MoE Transformer, large-scale motion data, and an optimized training-to-deployment pipeline, delivering stronger motion tracking with real-time inference efficiency.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/media/holomotion_overview.jpg\" alt=\"HoloMotion overview\" width=\"100%\"\u003e\n\u003c/p\u003e\n\n\n## Scales Toward 4-Any Humanoid Control\n\nThe roadmap of HoloMotion advances through four generalization targets, from motion imitation to command following, terrain adaptation, and embodiment transfer.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/media/holomotion_roadmap.jpg\" alt=\"HoloMotion 4-Any roadmap\" width=\"100%\"\u003e\n\u003c/p\u003e\n\n| Version  | Target Capability | Status      | Description                                                                                                                         |\n| -------- | ----------------- | ----------- | ----------------------------------------------------------------------------------------------------------------------------------- |\n| **v1.x** | Any Pose          | ✅ Done        | Achieve robust tracking and imitation of diverse, whole-body human motions, forming the core of the imitation learning capability.  |\n| **v2.x** | Any Command       | 🚀 Next        | Enable language- and task-conditioned motion generation, allowing for goal-directed and interactive behaviors.                      |\n| **v3.x** | Any Terrain       | 🧭 Planned    | Master adaptation to uneven, dynamic, and complex terrains, enhancing real-world operational robustness.                            |\n| **v4.x** | Any Embodiment    | 🧭 Planned    | Generalize control policies across humanoids with varying morphologies and kinematics, achieving true embodiment-level abstraction. |\n\n\n## Closes the Loop From Data to Robots\n\nHoloMotion provides a clear, modular framework for bridging motion data, policy learning, simulation evaluation, and real-robot deployment.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/media/holomotion_pipeline.jpg\" alt=\"HoloMotion pipeline\" width=\"100%\"\u003e\n\u003c/p\u003e\n\n## No Per-User Training Required\n\nWhether you want to replay motions, stream live teleoperation, or train a custom policy, HoloMotion provides a direct path into the workflow:\n\n| User Goal | Start Here | What You Need |\n| --------- | ---------- | ------------- |\n| **Offline motion tracking**\u003cbr/\u003eReplay local motion clips for demos such as dance or scripted performances. | [Real-world deployment: Offline Motion](docs/realworld_deployment.md#offline-motion) | A pretrained policy and retargeted `.npz` motion clips. No model training is required. |\n| **Online motion tracking**\u003cbr/\u003eFollow live VR or teleoperation motion streams. | [Real-world deployment: Online Motion](docs/realworld_deployment.md#online-motion) | A pretrained policy, robot deployment setup, and a live motion source. No model training is required. |\n| **Train your own model**\u003cbr/\u003eBuild a custom policy from your own motion data. | [Environment setup](docs/environment_setup.md) → [Data curation](docs/smpl_data_curation.md) → [Retargeting](docs/motion_retargeting.md) → [Training](docs/train_motion_tracking.md) → [Evaluation](docs/evaluate_motion_tracking.md) | Training environment, curated motion data, retargeted HDF5 datasets, and GPU resources. |\n\n# Join Us\n\nWe are hiring full-time engineers, new graduates, and interns who are excited about humanoid robots, motion control, and embodied intelligence.\nSend your resume by scanning the **WeChat** QR code below to get in touch with us.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg width=\"420\" height=\"150\" src=\"assets/media/qr_codes.jpg\" hspace=\"10\"\u003e\n\u003c/p\u003e\n\n# Citation\n\n```\n@misc{chen2026holomotion1,\n  title = {HoloMotion-1 Technical Report},\n  author = {Maiyue Chen and Kaihui Wang and Bo Zhang and Xihan Ma and Zhiyuan Yang and Yi Ren and Qijun Huang and Zihao Zhu and Yucheng Wang and Zhizhong Su},\n  year = {2026},\n  eprint = {2605.15336},\n  archivePrefix = {arXiv},\n  primaryClass = {cs.RO},\n  url = {https://arxiv.org/abs/2605.15336}\n}\n```\n\n# Acknowledgements\n\nThis project is built upon and inspired by several outstanding open source projects:\n\n- [ASAP](https://github.com/LeCAR-Lab/ASAP)\n- [BeyondMimic](https://github.com/HybridRobotics/whole_body_tracking)\n- [GMR](https://github.com/YanjieZe/GMR)\n- [GVHMR](https://github.com/zju3dv/GVHMR)\n- [Humanoidverse](https://github.com/LeCAR-Lab/HumanoidVerse)\n- [Mink](https://github.com/kevinzakka/mink)\n- [MotionMillion](https://github.com/VankouF/MotionMillion-Codes)\n- [PBHC](https://github.com/TeleHuman/PBHC)\n- [PHC](https://github.com/ZhengyiLuo/PHC?tab=readme-ov-file)\n- [ProtoMotion](https://github.com/NVlabs/ProtoMotions/tree/main/protomotions)\n- [SONIC](https://github.com/NVlabs/GR00T-WholeBodyControl)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhorizonrobotics%2Fholomotion","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhorizonrobotics%2Fholomotion","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhorizonrobotics%2Fholomotion/lists"}