Genie-Envisioner-v1.0

Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation

<div id="top" align="center">

<a href='https://arxiv.org/abs/2508.05635'><img src='https://img.shields.io/badge/arXiv-2508.05635-b31b1b.svg'></a>   <a href='https://genie-envisioner.github.io/'><img src='https://img.shields.io/badge/Site-GenieEnvisioner-blue'></a>  

</div>

Introduction

This repo provides the pretrained weights of Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation. You can refer to our Github Repo for more details.

Citation

If you find Genie Envisioner useful, please feel free to help ⭐ the </a> <a style='font-size:18px;color: #FF5DB0' href='https://github.com/AgibotTech/Genie-Envisioner'>[Github Repo]</a> or citing our paper.

Thanks!

@article{liao2025genie,
  title={Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation},
  author={Liao, Yue and Zhou, Pengfei and Huang, Siyuan and Yang, Donglin and Chen, Shengcong and Jiang, Yuxin and Hu, Yue and Cai, Jingbin and Liu, Si and Luo, Jianlan, Chen Liliang, Yan Shuicheng, Yao Maoqing, Ren Guanghui},
  journal={arXiv preprint arXiv:2508.05635},
  year={2025}
}

License

GE-base use the pretrained weights ltx-video-2b-v0.9, which means the checkpoint GE_base_fast_v0.1 and GE_base_slow_v0.1 is under the same license of ltx-video-2b-v0.9.

<!-- GE uses the weights of vae and tokenizer in [LTX-Video](https://huggingface.co/Lightricks/LTX-Video), which means the directories of vae and tokenizer weights are under [the same license](LTX-Video-Open-Weights-License-0.X.txt). -->

Other data and codes of this repo are under CC BY-NC-SA 4.0.