# Infinity $\infty$: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

<div align="center">

[![demo platform](https://img.shields.io/badge/Play%20with%20Infinity%21-Infinity%20demo%20platform-lightblue)](https://opensource.bytedance.com/gmpt/t2i/invite)&nbsp;
[![arXiv](https://img.shields.io/static/v1?label=Project%20Page&message=Github&color=blue&logo=github-pages)](https://foundationvision.github.io/infinity.project/)&nbsp;
[![arXiv](https://img.shields.io/badge/arXiv%20paper-2412.04431-b31b1b.svg)](https://arxiv.org/abs/2412.04431)&nbsp;
[![huggingface weights](https://img.shields.io/badge/%F0%9F%A4%97%20Weights-FoundationVision/Infinity-yellow)](https://huggingface.co/FoundationVision/infinity)&nbsp;
[![code](https://img.shields.io/badge/%F0%9F%A4%96%20Code-FoundationVision/Infinity-green)](https://github.com/FoundationVision/Infinity)&nbsp;

</div>
<p align="center" style="font-size: larger;">
  <a href="https://arxiv.org/abs/2412.04431">Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis</a>
</p>




## 📌 Note
This repo is used for hosting Infinity's checkpoints. For more details, please refer to [![code](https://img.shields.io/badge/%F0%9F%A4%96%20Code-FoundationVision/Infinity-green)](https://github.com/FoundationVision/Infinity)&nbsp;


## 📖 Introduction
We present Infinity, a Bitwise Visual AutoRegressive Modeling capable of generating high-resolution and photorealistic images. Infinity redefines visual autoregressive model under a bitwise token prediction framework with an infinite-vocabulary tokenizer & classifier and bitwise self-correction. Theoretically scaling the tokenizer vocabulary size to infinity and concurrently scaling the transformer size, our method significantly unleashes powerful scaling capabilities. Infinity sets a new record for autoregressive text-to-image models, outperforming top-tier diffusion models like SD3-Medium and SDXL. Notably, Infinity surpasses SD3-Medium by improving the GenEval benchmark score from 0.62 to 0.73 and the ImageReward benchmark score from 0.87 to 0.96, achieving a win rate of 66%. Without extra optimization, Infinity generates a high-quality 1024×1024 image in 0.8 seconds, making it 2.6× faster than SD3-Medium and establishing it as the fastest text-to-image model.


## 📀 Infinity Model ZOO
We provide Infinity models for you to play with, which are on <a href='https://huggingface.co/FoundationVision/infinity'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20weights-FoundationVision/Infinity-yellow'></a> or can be downloaded from the following links:

### Visual Tokenizer

|   vocabulary    | stride |   IN-256 rFID $\downarrow$    | IN-256 PSNR $\uparrow$ | IN-512 rFID $\downarrow$ | IN-512 PSNR $\uparrow$ | HF weights🤗                                                                        |
|:----------:|:-----:|:--------:|:---------:|:-------:|:-------:|:------------------------------------------------------------------------------------|
|  $V_d=2^{16}$   |  16  |   1.22   |  20.9   |    0.31    |  22.6   | [infinity_vae_d16.pth](https://huggingface.co/FoundationVision/infinity/blob/main/infinity_vae_d16.pth) |
|  $V_d=2^{24}$   |  16  |   0.75   |  22.0   |    0.30    |  23.5   | [infinity_vae_d24.pth](https://huggingface.co/FoundationVision/infinity/blob/main/infinity_vae_d24.pth) |
|  $V_d=2^{32}$   |  16  |   0.61   |  22.7   |    0.23    |  24.4   | [infinity_vae_d32.pth](https://huggingface.co/FoundationVision/infinity/blob/main/infinity_vae_d32.pth) |
|  $V_d=2^{64}$   |  16  |   0.33   |  24.9   |     0.15     |  26.4   | [infinity_vae_d64.pth](https://huggingface.co/FoundationVision/infinity/blob/main/infinity_vae_d64.pth) |
| $V_d=2^{32}$ |  16  | 0.75 |  21.9   |     0.32     |  23.6   | [infinity_vae_d32_reg.pth](https://huggingface.co/FoundationVision/infinity/blob/main/infinity_vae_d32_reg.pth) |

### Infinity
|   model    | Resolution |   GenEval    | DPG | HPSv2.1 | HF weights🤗                                                                        |
|:----------:|:-----:|:--------:|:---------:|:-------:|:------------------------------------------------------------------------------------|
|  Infinity-2B   |  1024  |   0.69 / 0.73 $^{\dagger}$   |    83.5    |  32.2   | [infinity_2b_reg.pth](https://huggingface.co/FoundationVision/infinity/blob/main/infinity_2b_reg.pth) |
|  Infinity-20B   |  1024  |  -  |    -    |  -   | [Coming Soon](TBD) |

${\dagger}$ result is tested with a [prompt rewriter](tools/prompt_rewriter.py). 

You can load these models to generate images via the codes in [interactive_infer.ipynb](tools/interactive_infer.ipynb). Note: you need to download [infinity_vae_d32reg.pth](https://huggingface.co/FoundationVision/infinity/blob/main/infinity_vae_d32_reg.pth) and [flan-t5-xl](https://huggingface.co/google/flan-t5-xl) first.



## 📖 Citation
If our work assists your research, feel free to give us a star ⭐ or cite us using:

```
@misc{han2024infinityscalingbitwiseautoregressive,
    title={Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis}, 
    author={Jian Han and Jinlai Liu and Yi Jiang and Bin Yan and Yuqi Zhang and Zehuan Yuan and Bingyue Peng and Xiaobing Liu},
    year={2024},
    eprint={2412.04431},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
    url={https://arxiv.org/abs/2412.04431}, 
}
```