ICLR 2026

MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer

Yanghao Li, Rui Qian, Bowen Pan, Haotian Zhang, Haoshuo Huang, Bowen Zhang, Jialing Tong, Haoxuan You, Xianzhi Du, Zhe Gan, Hyunjik Kim, Chao Jia, Zhenbang Wang, Yinfei Yang, Mingfei Gao, Zi-Yi Dou, Wenze Hu, Chang Gao, Dongxu Li, Philipp Dufter, Zirui Wang, Guoli Yin, Zhengdong Zhang, Chen Chen, Yang Zhao, Ruoming Pang, Zhifeng Chen

Paper ↗Project ↗Cite

A hybrid vision tokenizer brings visual understanding and image generation into one scalable multimodal model.

MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer — overview figure

BibTeX citation

@inproceedings{manzano2026,
  title = {MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer},
  author = {Yanghao Li and Rui Qian and Bowen Pan and Haotian Zhang and Haoshuo Huang and Bowen Zhang and Jialing Tong and Haoxuan You and Xianzhi Du and Zhe Gan and Hyunjik Kim and Chao Jia and Zhenbang Wang and Yinfei Yang and Mingfei Gao and Zi-Yi Dou and Wenze Hu and Chang Gao and Dongxu Li and Philipp Dufter and Zirui Wang and Guoli Yin and Zhengdong Zhang and Chen Chen and Yang Zhao and Ruoming Pang and Zhifeng Chen},
  booktitle = {ICLR},
  year = {2026},
  url = {https://arxiv.org/abs/2509.16197}
}