π WeTok: Powerful Discrete Tokenization for High-Fidelity Visual Reconstruction
[](https://arxiv.org/abs/2508.05599)
[](https://github.com/zhuangshaobin/WeTok)
[](https://huggingface.co/GrayShine/WeTok)
This project introduces **WeTok**, a powerful discrete visual tokenizer designed to resolve the long-standing conflict between compression efficiency and reconstruction fidelity. WeTok achieves state-of-the-art reconstruction quality, surpassing previous leading discrete and continuous tokenizers.
> WeTok: Powerful Discrete Tokenization for High-Fidelity Visual Reconstruction
> [Shaobin Zhuang](https://scholar.google.com/citations?user=PGaDirMAAAAJ&hl=zh-CN&oi=ao), [Yiwei Guo](https://scholar.google.com/citations?user=HCAyeJIAAAAJ&hl=zh-CN&oi=ao), [Canmiao Fu](), [Zhipeng Huang](), [Zeyue Tian](https://scholar.google.com/citations?user=dghq4MQAAAAJ&hl=zh-CN&oi=ao), [Ying Zhang](https://scholar.google.com/citations?user=R_psgxkAAAAJ&hl=zh-CN&oi=ao), [Chen Li](https://scholar.google.com/citations?hl=zh-CN&user=WDJL3gYAAAAJ), [Yali Wang](https://scholar.google.com/citations?hl=zh-CN&user=hD948dkAAAAJ)
> Shanghai Jiao Tong University, WeChat Vision (Tencent Inc.), Shenzhen Institutes of Advanced Technology (Chinese Academy of Sciences), Hong Kong University of Science and Technology, Shanghai AI Laboratory
> πWeTok.md
> ```
> @article{zhuang2026wetok,
> title={WeTok: Powerful Discrete Tokenization for High-Fidelity Visual Reconstruction},
> author={Zhuang, Shaobin and Guo, Yiwei and Fu, Canmiao and Huang, Zhipeng and Tian, Zeyue and Zhang, Ying and Li, Chen and Wang, Yali},
> journal={arXiv preprint arXiv:2508.05599},
> year={2025}
> }
> ```
WeTok achieves a new state-of-the-art in reconstruction fidelity, surpassing both discrete and continuous tokenizers, while offering high compression ratios.
## π° News
* **[2025.08.08]** π π π We are excited to release **WeTok**, a powerful discrete tokenizer featuring our novel **Group-Wise Lookup-Free Quantization (GQ)** and a **Generative Decoder (GD)**. Code and pretrained models are now available!
## π Implementations
### π οΈ Installation
- **Dependencies**:
```
bash env.sh
```
### Evaluation
- **Evaluation on ImageNet 50K Validation Set**
The dataset should be organized as follows:
```
imagenet
βββ val/
βββ ...
```
Run the 256Γ256 resolution evaluation script:
```
bash scripts/evaluation/imagenet_evaluation_256_dist.sh
```
Run the original resolution evaluation script:
```
bash scripts/evaluation/imagenet_evaluation_original_dist.sh
```
- **Evaluation on MS-COCO Val2017**
The dataset should be organized as follows:
```
MSCOCO2017
βββ val2017/
βββ ...
```
Run the evaluation script:
```
bash scripts/evaluation/mscocoval_evaluation_256_dist.sh
```
Run the original resolution evaluation script:
```
bash scripts/evaluation/mscoco_evaluation_original_dist.sh
```
### Inference
Simply test the effect of each model reconstruction:
```
bash scripts/inference/reconstruct_image.sh
```
Qualitative comparison of 512 Γ 512 image reconstruction on TokBench.
WeTok-AR-XL generated samples at 256 Γ 256 resolution.
## β€οΈ Acknowledgement
Our work builds upon the foundations laid by many excellent projects in the field. We would like to thank the authors of [Open-MAGVIT2](https://arxiv.org/abs/2409.04410). We also drew inspiration from the methodologies presented in [LFQ](https://arxiv.org/abs/2310.05737), [BSQ](https://arxiv.org/abs/2406.07548). We are grateful for their contributions to the community.