Neighboring Autoregressive Modeling for Efficient Visual Generation

   

Demo Video

🔥 Update

• [2025.03.17] C2i, t2i, c2v training and sampling code are released !

🌿 Introduction

We introduce NAR, a new "next-neighbor prediction" paradigm for efficient and high-quality visual generation. NAR achieves state-of-the-art generation quality and efficiency trade-off for both image and video generation tasks. All the training codes, data and models are open-sourced.

All pretrained models can be downloaded from HuggingFace.

BibTeX

@article{he2025nar,
  title={Neighboring Autoregressive Modeling for Efficient Visual Generation},
  author={He, Yefei and He, Yuanyu and He, Shaoxuan and Chen, Feng and Zhou, Hong and Zhang, Kaipeng and Zhuang, Bohan},
  journal={arXiv preprint arXiv:2503.10696},
  year={2025}
}