RoboDojo

# RoboDojo RoboDojo is a unified simulation-and-real-world benchmark for evaluating generalist robot manipulation policies. It provides benchmark assets, task configurations, and evaluation data for reproducible robot-manipulation research. The benchmark covers diverse manipulation scenarios across multiple robot embodiments, including simulation tasks and real-world evaluation tasks. It is designed to assess capabilities such as generalization, memory, precision, long-horizon planning, and open-world manipulation. ## Dataset Contents This dataset repository contains RoboDojo dataset files and associated metadata. Please browse the **Dataset Files** tab on ModelScope for the available files, directory structure, and file-level details. ## Intended Use RoboDojo is intended for: - Evaluating generalist robot manipulation policies - Developing and validating simulation-based manipulation systems - Reproducing RoboDojo benchmark experiments ## Related Resources - [RoboDojo Website](https://robodojo-benchmark.com/) - [Documentation](https://robodojo-benchmark.com/doc/) - [Benchmark Paper](https://arxiv.org/abs/2607.04434) - [Leaderboard](https://robodojo-benchmark.com/leaderboard) ## Download Refer to [install-and-download](https://robodojo-benchmark.com/doc/usage/install-and-download/) ## Citation If you use RoboDojo in your research, please cite: ```bibtex @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} } ``` ## License This dataset is released under the Apache License 2.0.