[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72512":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":30,"readmeContent":31,"aiSummary":32,"trendingCount":16,"starSnapshotCount":16,"syncStatus":33,"lastSyncTime":34,"discoverSource":35},72512,"RoboTwin","RoboTwin-Platform\u002FRoboTwin","RoboTwin-Platform","RoboTwin 2.0 Offical Repo","https:\u002F\u002Frobotwin-platform.github.io",null,"Python",2431,388,38,61,0,15,43,129,45,109.77,"MIT License",false,"main",[26,27,28,29],"benchmark","data-generator","embodied-ai","robotics","2026-06-12 04:01:06","\u003Ch1 align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Frobotwin-benchmark.github.io\">\u003Cb>RoboTwin\u003C\u002Fb> Bimanual Robotic Manipulation Platform\u003Cbr>\u003C\u002Fa>\n\u003C\u002Fh1>\n\u003Ch2 align=\"center\">Lastest Version: RoboTwin 2.0\u003Cbr>🤲 \u003Ca href=\"https:\u002F\u002Frobotwin-platform.github.io\u002F\">Webpage\u003C\u002Fa> | \u003Ca href=\"https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002F\">Document\u003C\u002Fa> | \u003Ca href=\"https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.18088\">Paper\u003C\u002Fa> | \u003Ca href=\"https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fcommunity\u002Findex.html\">Community\u003C\u002Fa> | \u003Ca href=\"https:\u002F\u002Frobotwin-platform.github.io\u002Fleaderboard\">Leaderboard\u003C\u002Fa>\u003C\u002Fh2>\n\nhttps:\u002F\u002Fprivate-user-images.githubusercontent.com\u002F88101805\u002F463126988-e3ba1575-4411-4a36-ad65-f0b2f49890c3.mp4\n\n**[2.0 Version (lastest)]** RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation\u003Cbr>\n\u003Ci>Under Review 2025\u003C\u002Fi>: [Webpage](https:\u002F\u002Frobotwin-platform.github.io\u002F) | [Document](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc) | [PDF](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2506.18088) | [arXiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.18088) | [Talk (in Chinese)](https:\u002F\u002Fwww.bilibili.com\u002Fvideo\u002FBV18p3izYE63\u002F?spm_id_from=333.337.search-card.all.click) | [机器之心](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002FSwORezmol2Qd9YdrGYchEA) | [Leaderboard](https:\u002F\u002Frobotwin-platform.github.io\u002Fleaderboard)\u003Cbr>\n> \u003Ca href=\"https:\u002F\u002Ftianxingchen.github.io\u002F\">Tianxing Chen\u003C\u002Fa>\u003Csup>\\*\u003C\u002Fsup>, Zanxin Chen\u003Csup>\\*\u003C\u002Fsup>, Baijun Chen\u003Csup>\\*\u003C\u002Fsup>, Zijian Cai\u003Csup>\\*\u003C\u002Fsup>, \u003Ca href=\"https:\u002F\u002F10-oasis-01.github.io\">Yibin Liu\u003C\u002Fa>\u003Csup>\\*\u003C\u002Fsup>, \u003Ca href=\"https:\u002F\u002Fkolakivy.github.io\u002F\">Qiwei Liang\u003C\u002Fa>, Zixuan Li, Xianliang Lin, \u003Ca href=\"https:\u002F\u002Fgeyiheng.github.io\">Yiheng Ge\u003C\u002Fa>, Zhenyu Gu, Weiliang Deng, Yubin Guo, Tian Nian, Xuanbing Xie, \u003Ca href=\"https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fyusen-qin-5b23345b\u002F\">Qiangyu Chen\u003C\u002Fa>, Kailun Su, Tianling Xu, \u003Ca href=\"http:\u002F\u002Fluoping.me\u002F\">Guodong Liu\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Faaron617.github.io\u002F\">Mengkang Hu\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fc7w.tech\u002Fabout\">Huan-ang Gao\u003C\u002Fa>, Kaixuan Wang, \u003Ca href=\"https:\u002F\u002Fliang-zx.github.io\u002F\">Zhixuan Liang\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fyusen-qin-5b23345b\u002F\">Yusen Qin\u003C\u002Fa>, Xiaokang Yang, \u003Ca href=\"http:\u002F\u002Fluoping.me\u002F\">Ping Luo\u003C\u002Fa>\u003Csup>†\u003C\u002Fsup>, \u003Ca href=\"https:\u002F\u002Fyaomarkmu.github.io\u002F\">Yao Mu\u003C\u002Fa>\u003Csup>†\u003C\u002Fsup>\n\n**[RoboTwin Dual-Arm Collaboration Challenge@CVPR'25 MEIS Workshop]** RoboTwin Dual-Arm Collaboration Challenge Technical Report at CVPR 2025 MEIS Workshop\u003Cbr>\nOfficial Technical Report: [PDF](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2506.23351) | [arXiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.23351) | [量子位](https:\u002F\u002Fmp.weixin.qq.com\u002Fs\u002Fqxqs9vvvHsAJ-0hoYANYzQ)\u003Cbr>\n\n**[1.0 Version]** RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins\u003Cbr>\nAccepted to \u003Ci style=\"color: red; display: inline;\">\u003Cb>CVPR 2025 (Highlight)\u003C\u002Fb>\u003C\u002Fi>: [PDF](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2504.13059) | [arXiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.13059)\u003Cbr>\n> \u003Ca href=\"https:\u002F\u002Fyaomarkmu.github.io\u002F\">Yao Mu\u003C\u002Fa>\u003Csup>* †\u003C\u002Fsup>, \u003Ca href=\"https:\u002F\u002Ftianxingchen.github.io\">Tianxing Chen\u003C\u002Fa>\u003Csup>* \u003C\u002Fsup>, Zanxin Chen\u003Csup>* \u003C\u002Fsup>, \u003Ca href=\"https:\u002F\u002Fshijiapeng03.github.io\">Shijia Peng\u003C\u002Fa>\u003Csup>* \u003C\u002Fsup>, Zhiqian Lan, Zeyu Gao, Zhixuan Liang, Qiaojun Yu, Yude Zou, Mingkun Xu, Lunkai Lin, Zhiqiang Xie, Mingyu Ding, \u003Ca href=\"http:\u002F\u002Fluoping.me\u002F\">Ping Luo\u003C\u002Fa>\u003Csup>†\u003C\u002Fsup>.\n\n**[Early Version]** RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins (early version)\u003Cbr>\nAccepted to \u003Ci style=\"color: red; display: inline;\">\u003Cb>ECCV Workshop 2024 (Best Paper Award)\u003C\u002Fb>\u003C\u002Fi>: [PDF](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2409.02920) | [arXiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.02920)\u003Cbr>\n> \u003Ca href=\"https:\u002F\u002Fyaomarkmu.github.io\u002F\">Yao Mu\u003C\u002Fa>\u003Csup>* †\u003C\u002Fsup>, \u003Ca href=\"https:\u002F\u002Ftianxingchen.github.io\">Tianxing Chen\u003C\u002Fa>\u003Csup>* \u003C\u002Fsup>, Shijia Peng\u003Csup>*\u003C\u002Fsup>, Zanxin Chen\u003Csup>*\u003C\u002Fsup>, Zeyu Gao, Zhiqian Lan, Yude Zou, Lunkai Lin, Zhiqiang Xie, \u003Ca href=\"http:\u002F\u002Fluoping.me\u002F\">Ping Luo\u003C\u002Fa>\u003Csup>†\u003C\u002Fsup>.\n\n\n\n# 📚 Overview\n\n| Branch Name | Link |\n|-------------|------|\n| 2.0 Version Branch | [main](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002Fmain) (latest) |\n| IsaacLab-Arena Branch | [IsaacLab-Arena](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FIsaacLab-Arena) |\n| RLinf Branch | [RLinf_support](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FRLinf_support) |\n| WBCD 2026 Branch | [WBCD-2026](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FWBCD-2026) |\n| 1.0 Version Branch | [1.0 Version](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FRoboTwin-1.0) |\n| 1.0 Version Code Generation Branch | [1.0 Version GPT](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002Fgpt) |\n| Early Version Branch | [Early Version](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002Fearly_version) |\n| 第十九届“挑战杯”人工智能专项赛分支 | [Challenge-Cup-2025](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FChallenge-Cup-2025) |\n| CVPR 2025 Challenge Round 1 Branch | [CVPR-Challenge-2025-Round1](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FCVPR-Challenge-2025-Round1) |\n| CVPR 2025 Challenge Round 2 Branch | [CVPR-Challenge-2025-Round2](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FCVPR-Challenge-2025-Round2) |\n\n\n\n# 🐣 Update\n* **2026\u002F03\u002F03**, We release [RMBench](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRMBench), which is a memory-dependent manipulation benchmark built upon RoboTwin 2.0.\n* **2026\u002F02\u002F20**, Usage supported in \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FstarVLA\u002FstarVLA\">StarVLA\u003C\u002Fa>, which is a user-friendly codebase for VLA development.\n* **2026\u002F01\u002F23**, We update IsaacLab-Arena and \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FRLinf\u002FRLinf\">RLinf\u003C\u002Fa> support (contributed by RLinf team).\n* **2025\u002F08\u002F28**, We update the RoboTwin 2.0 Paper [PDF](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2506.18088).\n* **2025\u002F08\u002F25**, We fix ACT deployment code and update the [leaderboard](https:\u002F\u002Frobotwin-platform.github.io\u002Fleaderboard).\n* **2025\u002F08\u002F06**, We release RoboTwin 2.0 Leaderboard: [leaderboard website](https:\u002F\u002Frobotwin-platform.github.io\u002Fleaderboard).\n* **2025\u002F07\u002F23**, RoboTwin 2.0 received Outstanding Poster at ChinaSI 2025 (Ranking 1st).\n* **2025\u002F07\u002F19**, We Fix DP3 evaluation code error. We will update RoboTwin 2.0 paper next week.\n* **2025\u002F07\u002F09**, We update endpose control mode, please see [[RoboTwin Doc - Usage - Control Robot](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002Fcontrol-robot.html)] for more details.\n* **2025\u002F07\u002F08**, We upload [Challenge-Cup-2025](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Ftree\u002FChallenge-Cup-2025) Branch (第十九届挑战杯分支).\n* **2025\u002F07\u002F02**, Fix Piper Wrist Bug [[issue](https:\u002F\u002Fgithub.com\u002FRoboTwin-Platform\u002FRoboTwin\u002Fissues\u002F104)]. Please redownload the embodiment asset.\n* **2025\u002F07\u002F01**, We release Technical Report of RoboTwin Dual-Arm Collaboration Challenge @ CVPR 2025 MEIS Workshop [[arXiv](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.23351)] !\n* **2025\u002F06\u002F21**, We release RoboTwin 2.0 [[Webpage](https:\u002F\u002Frobotwin-platform.github.io\u002F)] !\n* **2025\u002F04\u002F11**, RoboTwin is seclected as \u003Ci>CVPR Highlight paper\u003C\u002Fi>!\n* **2025\u002F02\u002F27**, RoboTwin is accepted to \u003Ci>CVPR 2025\u003C\u002Fi> ! \n* **2024\u002F09\u002F30**, RoboTwin (Early Version) received \u003Ci>the Best Paper Award  at the ECCV Workshop\u003C\u002Fi>!\n* **2024\u002F09\u002F20**, Officially released RoboTwin.\n\n# 🛠️ Installation\n\nSee [RoboTwin 2.0 Document (Usage - Install & Download)](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002Frobotwin-install.html) for installation instructions. It takes about 20 minutes for installation.\n\n# 🤷‍♂️ Tasks Informations\nSee [RoboTwin 2.0 Tasks Doc](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Ftasks\u002Findex.html) for more details.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\".\u002Fassets\u002Ffiles\u002F50_tasks.gif\" width=\"100%\">\n\u003C\u002Fp>\n\n# 🧑🏻‍💻 Usage \n\n## Document\n\n> Please Refer to [RoboTwin 2.0 Document (Usage)](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002Findex.html) for more details.\n\n## Data Collection\nWe provide over 100,000 pre-collected trajectories as part of the open-source release [RoboTwin Dataset](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FTianxingChen\u002FRoboTwin2.0\u002Ftree\u002Fmain\u002Fdataset).\nHowever, we strongly recommend users to perform data collection themselves due to the high configurability and diversity of task and embodiment setups.\n\n\u003Cimg src=\".\u002Fassets\u002Ffiles\u002Fdomain_randomization.png\" alt=\"description\" style=\"display: block; margin: auto; width: 100%;\">\n\n## 1. Task Running and Data Collection\nRunning the following command will first search for a random seed for the target collection quantity, and then replay the seed to collect data.\n\n```\nbash collect_data.sh ${task_name} ${task_config} ${gpu_id}\n# Example: bash collect_data.sh beat_block_hammer demo_randomized 0\n```\n\n## 2. Modify Task Config\n☝️ See [RoboTwin 2.0 Tasks Configurations Doc](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002Fconfigurations.html) for more details.\n\n# 🚴‍♂️ Policy Baselines\n## Policies Support\n[DP](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FDP.html), [ACT](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FACT.html), [DP3](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FDP3.html), [RDT](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FRDT.html), [PI0](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FPi0.html), [OpenVLA-oft](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FOpenVLA-oft.html)\n\n[TinyVLA](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FTinyVLA.html), [DexVLA](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FDexVLA.html) (Contributed by Media Group)\n\n[LLaVA-VLA](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FLLaVA-VLA.html) (Contributed by IRPN Lab, HKUST(GZ))\n\n[GO-1](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002FGO1.html) (Contributed by GO-1 Team)\n\nDeploy Your Policy: [Guidance](https:\u002F\u002Frobotwin-platform.github.io\u002Fdoc\u002Fusage\u002Fdeploy-your-policy.html)\n\n⏰ TODO: G3Flow, HybridVLA, SmolVLA, AVR, UniVLA\n\n# 🏄‍♂️ Experiment & LeaderBoard\n\n> We recommend that the RoboTwin Platform can be used to explore the following topics: \n> 1. single - task fine - tuning capability\n> 2. visual robustness\n> 3. language diversity robustness (language condition)\n> 4. multi-tasks capability\n> 5. cross-embodiment performance\n\nThe full leaderboard and setting can be found in: [https:\u002F\u002Frobotwin-platform.github.io\u002Fleaderboard](https:\u002F\u002Frobotwin-platform.github.io\u002Fleaderboard).\n\n# 💽 Pre-collected Large-scale Dataset\n\nPlease refer to [RoboTwin 2.0 Dataset - Huggingface](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FTianxingChen\u002FRoboTwin2.0\u002Ftree\u002Fmain\u002Fdataset).\n\n# 👍 Citations\nIf you find our work useful, please consider citing:\n\n\u003Cb>RoboTwin 2.0\u003C\u002Fb>: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation\n```\n@article{chen2025robotwin,\n  title={Robotwin 2.0: A scalable data generator and benchmark with strong domain randomization for robust bimanual robotic manipulation},\n  author={Chen, Tianxing and Chen, Zanxin and Chen, Baijun and Cai, Zijian and Liu, Yibin and Li, Zixuan and Liang, Qiwei and Lin, Xianliang and Ge, Yiheng and Gu, Zhenyu and others},\n  journal={arXiv preprint arXiv:2506.18088},\n  year={2025}\n}\n```\n\n\u003Cb>RoboTwin\u003C\u002Fb>: Dual-Arm Robot Benchmark with Generative Digital Twins, accepted to \u003Ci style=\"color: red; display: inline;\">\u003Cb>CVPR 2025 (Highlight)\u003C\u002Fb>\u003C\u002Fi>\n```\n@InProceedings{Mu_2025_CVPR,\n    author    = {Mu, Yao and Chen, Tianxing and Chen, Zanxin and Peng, Shijia and Lan, Zhiqian and Gao, Zeyu and Liang, Zhixuan and Yu, Qiaojun and Zou, Yude and Xu, Mingkun and Lin, Lunkai and Xie, Zhiqiang and Ding, Mingyu and Luo, Ping},\n    title     = {RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins},\n    booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR)},\n    month     = {June},\n    year      = {2025},\n    pages     = {27649-27660}\n}\n```\n\nBenchmarking Generalizable Bimanual Manipulation: RoboTwin Dual-Arm Collaboration Challenge at CVPR 2025 MEIS Workshop\n```\n@article{chen2025benchmarking,\n  title={Benchmarking Generalizable Bimanual Manipulation: RoboTwin Dual-Arm Collaboration Challenge at CVPR 2025 MEIS Workshop},\n  author={Chen, Tianxing and Wang, Kaixuan and Yang, Zhaohui and Zhang, Yuhao and Chen, Zanxin and Chen, Baijun and Dong, Wanxi and Liu, Ziyuan and Chen, Dong and Yang, Tianshuo and others},\n  journal={arXiv preprint arXiv:2506.23351},\n  year={2025}\n}\n```\n\n\u003Cb>RoboTwin\u003C\u002Fb>: Dual-Arm Robot Benchmark with Generative Digital Twins (early version), accepted to \u003Ci style=\"color: red; display: inline;\">\u003Cb>ECCV Workshop 2024 (Best Paper Award)\u003C\u002Fb>\u003C\u002Fi>\n```\n@article{mu2024robotwin,\n  title={RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins (early version)},\n  author={Mu, Yao and Chen, Tianxing and Peng, Shijia and Chen, Zanxin and Gao, Zeyu and Zou, Yude and Lin, Lunkai and Xie, Zhiqiang and Luo, Ping},\n  journal={arXiv preprint arXiv:2409.02920},\n  year={2024}\n}\n```\n\n# 😺 Acknowledgement\n\n**Software Support**: D-Robotics, **Hardware Support**: AgileX Robotics, **AIGC Support**: Deemos.\n\nContact [Tianxing Chen](https:\u002F\u002Ftianxingchen.github.io) if you have any questions or suggestions.\n\n# 🏷️ License\nThis repository is released under the MIT license. See [LICENSE](.\u002FLICENSE) for additional details.\n","RoboTwin 是一个专注于双臂机器人操作的平台，旨在通过强大的领域随机化技术生成大规模数据集并提供基准测试。其核心功能包括可扩展的数据生成器、高度灵活的仿真环境以及对复杂任务的支持，使用Python语言开发。该平台适用于研究机构、高校及企业中的机器人技术研发人员，特别是那些致力于提高双臂协作机器人在真实世界中执行任务能力的研究者。通过提供详尽的文档、活跃的社区支持和公开的排行榜，RoboTwin促进了相关领域的创新与发展。",2,"2026-06-11 03:42:21","high_star"]