A Unified Sim-and-Real Benchmark for Comprehensive Evaluation of Generalist Robot Manipulation Policies

42 simulation tasks · 18 real-world tasks · 5 capability axes · 30 policies on the public leaderboard

RoboDojo evaluates generalist robot manipulation across simulation and the real world. Simulation covers generalization, memory, precision, long-horizon execution, and open-vocabulary instruction following. Real-world tests run on ARX X5, Piper, and Piper X through RoboDojo-RealEval. XPolicyLab is the shared policy stack: integrate once, then evaluate in sim and on the robot.

📦 Hugging Face Datasets

• 🗂️ RoboDojo: Official simulation and real-world benchmark data (HDF5 / LeRobot / depth exports). Apache-2.0.
• 🧩 GOAI-2026: Challenge assets released with the RoboDojo organization.

🔗 Official Links

• Website / docs: robodojo-benchmark.com · Wiki
• Paper: arXiv:2607.04434
• Benchmark repo: robodojo-benchmark/RoboDojo
• Policy stack: XPolicyLab/XPolicyLab
• Leaderboard: View leaderboard
• Contact: RoboDojoCommittee@gmail.com
• Community: WeChat · Discord · Feishu · QQ — see the project page

📚 Citation

If you use RoboDojo, XPolicyLab, or the leaderboard in your work, please cite:

@article{chen2026robodojo,
  title={{RoboDojo}: A Unified Sim-and-Real Benchmark for Comprehensive Evaluation of Generalist Robot Manipulation Policies},
  author={Chen, Tianxing and Chen, Yue and Li, Zixuan and Tang, Junyuan and Su, Kailun and Wan, Weijie and Chen, Baijun and Lu, Haoran and Yan, Haowen and Su, Honghao and others},
  journal={arXiv preprint arXiv:2607.04434},
  year={2026}
}

@article{community2026xpolicylab,
  title={{XPolicyLab}: A Unified Standard and Open Ecosystem for Robot Policy Evaluation and Deployment},
  author={Community, XPolicyLab and Chen, Tianxing and Chen, Yue and Nian, Tian and Cai, Zijian and Chen, Guangyu and Lin, Wenwei and Liang, Qiwei and Xiang, Peicheng and Su, Kailun and others},
  journal={arXiv preprint arXiv:2608.09892},
  year={2026}
}