Improve model card: Add metadata and external links
#1
by
nielsr
HF Staff
- opened
README.md
CHANGED
|
@@ -3,10 +3,17 @@ license: apache-2.0
|
|
| 3 |
tags:
|
| 4 |
- NextStep
|
| 5 |
- Image Tokenizer
|
|
|
|
|
|
|
| 6 |
---
|
|
|
|
| 7 |
# Improved Image Tokenizer
|
| 8 |
|
| 9 |
-
This is an improved image tokenizer of NextStep-1,
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
## Usage
|
| 12 |
|
|
@@ -78,4 +85,22 @@ quantitative metrics (rFID↓, PSNR↑, and SSIM↑) versus noise intensity. The
|
|
| 78 |
|
| 79 |
<div align='center'>
|
| 80 |
<img src="assets/robustness.png" class="interpolation-image" alt="arch." width="100%" />
|
| 81 |
-
</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
tags:
|
| 4 |
- NextStep
|
| 5 |
- Image Tokenizer
|
| 6 |
+
pipeline_tag: image-feature-extraction
|
| 7 |
+
library_name: diffusers
|
| 8 |
---
|
| 9 |
+
|
| 10 |
# Improved Image Tokenizer
|
| 11 |
|
| 12 |
+
This is an improved image tokenizer of NextStep-1, featuring a fine-tuned decoder with a frozen encoder. This model is based on the paper [NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale](https://huggingface.co/papers/2508.10711). The decoder refinement **improves performance** while preserving robust reconstruction quality. We **recommend using this Image Tokenizer** for optimal results with NextStep-1 models.
|
| 13 |
+
|
| 14 |
+
### Project Resources
|
| 15 |
+
* **Website:** [https://stepfun.ai/research/en/nextstep1](https://stepfun.ai/research/en/nextstep1)
|
| 16 |
+
* **Code:** [https://github.com/stepfun-ai/NextStep-1](https://github.com/stepfun-ai/NextStep-1)
|
| 17 |
|
| 18 |
## Usage
|
| 19 |
|
|
|
|
| 85 |
|
| 86 |
<div align='center'>
|
| 87 |
<img src="assets/robustness.png" class="interpolation-image" alt="arch." width="100%" />
|
| 88 |
+
</div>
|
| 89 |
+
|
| 90 |
+
## Acknowledgments
|
| 91 |
+
|
| 92 |
+
We would like to express our sincere thanks to theWe would like to sincerely thank Tianhong Li and Yonglong Tian for their
|
| 93 |
+
insightful discussions.
|
| 94 |
+
|
| 95 |
+
## Citation
|
| 96 |
+
|
| 97 |
+
If you find NextStep useful for your research and applications, please consider starring this repository and citing:
|
| 98 |
+
|
| 99 |
+
```bibtex
|
| 100 |
+
@article{nextstepteam2025nextstep1,
|
| 101 |
+
title={NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale},
|
| 102 |
+
author={NextStep Team and Chunrui Han and Guopeng Li and Jingwei Wu and Quan Sun and Yan Cai and Yuang Peng and Zheng Ge and Deyu Zhou and Haomiao Tang and Hongyu Zhou and Kenkun Liu and Ailin Huang and Bin Wang and Changxin Miao and Deshan Sun and En Yu and Fukun Yin and Gang Yu and Hao Nie and Haoran Lv and Hanpeng Hu and Jia Wang and Jian Zhou and Jianjian Sun and Kaijun Tan and Kang An and Kangheng Lin and Liang Zhao and Mei Chen and Peng Xing and Rui Wang and Shiyu Liu and Shutao Xia and Tianhao You and Wei Ji and Xianfang Zeng and Xin Han and Xuelin Zhang and Yana Wei and Yanming Xu and Yimin Jiang and Yingming Wang and Yu Zhou and Yucheng Han and Ziyang Meng and Binxing Jiao and Daxin Jiang and Xiangyu Zhang and Yibo Zhu},
|
| 103 |
+
journal={arXiv preprint arXiv:2508.10711},
|
| 104 |
+
year={2025}
|
| 105 |
+
}
|
| 106 |
+
```
|