File size: 14,595 Bytes

710296c

---
license: mit
pipeline_tag: any-to-any
library_name: transformers
---

<div align="center">
<h1>UniTok: A Unified Tokenizer <br> for Visual Generation and Understanding</h1>

[**Chuofan Ma**](https://machuofan.github.io/)<sup>1,2</sup> · [**Yi Jiang**](https://enjoyyi.github.io/)<sup>2&dagger;</sup> · [**Junfeng Wu**](https://wjf5203.github.io/)<sup>2,3</sup> · [**Jihan Yang**](https://jihanyang.github.io/)<sup>1</sup>
<br>
[**Xin Yu**](https://xinyu-andy.github.io/)<sup>1</sup> · [**Zehuan Yuan**](https://shallowyuan.github.io/)<sup>2*</sup> · [**Bingyue Peng**](https://openreview.net/profile?id=~BINGYUE_PENG1)<sup>2</sup> · [**Xiaojuan Qi**](https://xjqi.github.io/)<sup>1&dagger;*</sup>

<sup>1</sup>HKU&emsp;&emsp;&emsp;<sup>2</sup>ByteDance&emsp;&emsp;&emsp;<sup>3</sup>HUST
<br>
&dagger;project lead&emsp;&emsp;&emsp;*corresponding author

<a href="https://huggingface.co/papers/2502.20321"><img src='https://img.shields.io/badge/Paper-UniTok-red' alt='Paper PDF'></a>
<a href="https://foundationvision.github.io/UniTok/"><img src='https://img.shields.io/badge/Project_Page-UniTok-green' alt='Project Page'></a>
<a href="https://github.com/foundationvision/unitok"><img src='https://img.shields.io/badge/GitHub-Code-blue'></a>
<a href="https://huggingface.co/FoundationVision/unitok_tokenizer"><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Model-blue'></a>
<a href="https://huggingface.co/spaces/FoundationVision/UniTok"><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Demo-yellow'></a>
</div>

This repository implements UniTok, a unified visual tokenizer well-suited for both generation and understanding tasks.
It is compatible with autoregressive generative models (e.g. LlamaGen),
multimodal understanding models (e.g. LLaVA), and unified MLLMs (e.g. Chameleon and Liquid).

![teaser](https://github.com/FoundationVision/UniTok/raw/main/assets/teaser.png)

Built upon UniTok, we construct an MLLM capable of both multimodal generation and understanding
with the [Liquid](https://github.com/FoundationVision/Liquid/) framework,
which sets a new state-of-the-art among unified autoregressive MLLMs.

![teaser](https://github.com/FoundationVision/UniTok/raw/main/assets/samples.png)

## Abstract
Visual generative and understanding models typically rely on distinct tokenizers to process images, presenting a key challenge for unifying them within a single framework. Recent studies attempt to address this by connecting the training of VQVAE (for autoregressive generation) and CLIP (for understanding) to build a unified tokenizer. However, directly combining these training objectives has been observed to cause severe loss conflicts. In this paper, we show that reconstruction and semantic supervision do not inherently conflict. Instead, the underlying bottleneck stems from limited representational capacity of discrete token space. Building on these insights, we introduce UniTok, a unified tokenizer featuring a novel multi-codebook quantization mechanism that effectively scales up the vocabulary size and bottleneck dimension. In terms of final performance, UniTok sets a new record of 0.38 rFID and 78.6% zero-shot accuracy on ImageNet. Besides, UniTok can be seamlessly integrated into MLLMs to unlock native visual generation capability, without compromising the understanding performance. Additionally, we show that UniTok favors cfg-free generation, reducing gFID from 14.6 to 2.5 on ImageNet 256$\times$256 benchmark. GitHub: this https URL .

## News
**2025-09-18:** UniTok is accepted at NeurIPS 2025 as a spotlight.

**2025-05-19:** We find UniTok favors generation **without classifier-free-guidance** --
it reduces gFID (without cfg) from 14.6 to 2.51 on ImageNet 256x256 using LlamaGen-XXL as the generator.
Please refer to the updated [EVAL.md](https://github.com/FoundationVision/UniTok/blob/main/eval/EVAL.md) for more details.

**2025-04-15:** The [gradio demo](https://huggingface.co/spaces/FoundationVision/UniTok) of UniTok MLLM is available on Huggingface now!

**2025-04-02:** A new [checkpoint](https://huggingface.co/FoundationVision/unitok_tokenizer/tree/main)
of UniTok is released, which has better downstream task performance
by replacing the causal attention projection layer with full attention.
The [model weights](https://huggingface.co/FoundationVision/unitok_mllm)
of our unified MLLM are also available on Huggingface now!

**2025-02-28:** Paper, code, model, and [project page](https://foundationvision.github.io/UniTok/) for UniTok are all released.

## Performance

<table>
    <thead>
        <tr>
            <th>Method</th>
            <th>#Tokens</th>
            <th>rFID &darr;</th>
            <th>Accuracy</th>
        </tr>
    </thead>
    <tbody>
        <tr>
            <td colspan="4"><i>VQVAE Model</i></td>
        </tr>
        <tr align="center">
            <td>VQ-GAN</td>
            <td>256</td>
            <td>4.98</td>
            <td>--</td>
        </tr>
        <tr align="center">
            <td>RQ-VAE</td>
            <td>256</td>
            <td>1.30</td>
            <td>--</td>
        </tr>
        <tr align="center">
            <td>VAR</td>
            <td>680</td>
            <td>0.90</td>
            <td>--</td>
        </tr>
        <tr>
            <td colspan="4"><i>CLIP Model</i></td>
        </tr>
        <tr align="center">
            <td>CLIP</td>
            <td>256</td>
            <td>--</td>
            <td>76.2</td>
        </tr>
        <tr align="center">
            <td>SigLIP</td>
            <td>256</td>
            <td>--</td>
            <td>80.5</td>
        </tr>
        <tr align="center">
            <td>ViTamin</td>
            <td>256</td>
            <td>--</td>
            <td>81.2</td>
        </tr>
        <tr>
            <td colspan="4"><i>Unified Model</i></td>
        </tr>
        <tr align="center">
            <td>TokenFlow &dagger;</td>
            <td>680</td>
            <td>1.37</td>
            <td>--</td>
        </tr>
        <tr align="center">
            <td>VILA-U &dagger;</td>
            <td>256</td>
            <td>1.80</td>
            <td>73.3</td>
        </tr>
        <tr align="center">
            <td>UniTok</td>
            <td>256</td>
            <td>0.41</td>
            <td>70.8</td>
        </tr>
        <tr align="center">
            <td>UniTok &dagger;</td>
            <td>256</td>
            <td>0.38</td>
            <td>78.6</td>
        </tr>
    </tbody>
</table>


&dagger; indicates the model uses pretrained CLIP weights for initialization. Although CLIP weight initialization boosts ImageNet zero-shot accuracy,
we notice that random initialization leads to better downstream understanding performance.
We thus release the model checkpoint of UniTok that is trained from scratch.

## Model Weights

|    Model     | Res. | #Token |        Code Shape         | rFID |  Checkpoint  |
|:------------:|:----:|:------:|:-------------------------:|:----:|:------------:|
| UniTok-Large | 256  |  256   | 16 $\times$ 16 $\times$ 8 | 0.41 | [Download](https://huggingface.co/FoundationVision/unitok_tokenizer/blob/main/unitok_tokenizer.pth) |

## Usage

### Requirements
- Python ≥ 3.10
- PyTorch ≥ 2.3.1

### Installation

```bash
git clone https://github.com/FoundationVision/UniTok.git
cd UniTok
pip install -r requirements.txt
```

### Inference

Please download the [checkpoint](https://huggingface.co/FoundationVision/unitok_tokenizer) and fill in the `ckpt_path`.
```bash
python inference.py \
    --ckpt_path /path/to/unitok_tokenizer.pth \
    --src_img /path/to/test_img --rec_img /path/to/rec_img
```

### Unified MLLM Inference

The simplest code for Lumina-mGPT inference:

```python
from inference_solver import FlexARInferenceSolver
from PIL import Image

# ******************** Image Generation ********************
inference_solver = FlexARInferenceSolver(
    model_path="Alpha-VLLM/Lumina-mGPT-7B-768",
    precision="bf16",
    target_size=768,
)

q1 = f"Generate an image of 768x768 according to the following prompt:
" \
     f"Image of a dog playing water, and a waterfall is in the background."

# generated: tuple of (generated response, list of generated images)
generated = inference_solver.generate(
    images=[],
    qas=[[q1, None]],
    max_gen_len=8192,
    temperature=1.0,
    logits_processor=inference_solver.create_logits_processor(cfg=4.0, image_top_k=2000),
)

a1, new_image = generated[0], generated[1][0]


# ******************* Image Understanding ******************
inference_solver = FlexARInferenceSolver(
    model_path="Alpha-VLLM/Lumina-mGPT-7B-512",
    precision="bf16",
    target_size=512,
)

# "<|image|>" symbol will be replaced with sequence of image tokens before fed to LLM
q1 = "Describe the image in detail. <|image|>"

images = [Image.open("image.png")]
qas = [[q1, None]]

# `len(images)` should be equal to the number of appearance of "<|image|>" in qas
generated = inference_solver.generate(
    images=images,
    qas=qas,
    max_gen_len=8192,
    temperature=1.0,
    logits_processor=inference_solver.create_logits_processor(cfg=4.0, image_top_k=2000),
)

a1 = generated[0]
# generated[1], namely the list of newly generated images, should typically be empty in this case.


# ********************* Omni-Potent *********************
inference_solver = FlexARInferenceSolver(
    model_path="Alpha-VLLM/Lumina-mGPT-7B-768-Omni",
    precision="bf16",
    target_size=768,
)

# Example: Depth Estimation
# For more instructions, see demos/demo_image2image.py
q1 = "Depth estimation. <|image|>"
images = [Image.open("image.png")]
qas = [[q1, None]]

generated = inference_solver.generate(
    images=images,
    qas=qas,
    max_gen_len=8192,
    temperature=1.0,
    logits_processor=inference_solver.create_logits_processor(cfg=1.0, image_top_k=200),
)

a1 = generated[0]
new_image = generated[1][0]
```

### Training

- We train UniTok on [DataComp-1B](https://github.com/mlfoundations/datacomp).
Please follow the [instructions](https://github.com/mlfoundations/datacomp?tab=readme-ov-file#downloading-datacomp-1b) to download and prepare the data.

- Download the [models](https://huggingface.co/FoundationVision/unitok_external) used for loss calculation and put them under `./external`.

- Download the [ImageNet validation set](https://www.image-net.org/) for zero-shot accuracy evaluation.

- Download the ImageNet 256$\times$256 [reference batch](https://huggingface.co/datasets/FoundationVision/imagenet_reference_batch) for FID evaluation.

Configure `nnodes, nproc_per_node, node_rank, master_addr, master_port` in `launch.sh` and run:

```bash
bash launch.sh \
    --output_dir '/path/to/save/checkpoints/' \
    --train_data '/path/to/datacomp/shards/{00000000..00140146}.tar' \
    --imagenet_val '/path/to/imagenet_val/' \
    --fid_eval_src '/path/to/imagenet_reference_batch' \
    --fid_eval_dst '/path/to/save/imagenet_reconstructed_batch'
```
**Note:** For more hyper-parameter configurations, please check `utils/config.py`.

### Unified MLLM
We show that UniTok significantly boosts the performance of unified MLLMs.

Visual Understanding Performance on VQA Benchmarks.

|   Method   |      LLM       |  Res.   |  VQAv2   |   GQA    | TextVQA  |   POPE   |   MME    |  MM-Vet  |
|:----------:|:--------------:|:-------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|
|   Show-o   |  Phi-1.5-1.3B  |   256   |   59.3   |   48.7   |    -     |   73.8   |   948    |    -     |
|   Liquid   |    Gemma-7B    |   512   |   71.3   |   58.4   |   42.4   |   81.1   |   1119   |    -     |
|   VILA-U   |   Llama-2-7B   |   256   |   75.3   |   58.3   |   48.3   |   83.9   |   1336   |   27.7   |
| **UniTok** | **Llama-2-7B** | **256** | **76.8** | **61.1** | **51.6** | **83.2** | **1448** | **33.9** |

Visual Generation Performance on GenAI-Bench.

<table>
    <thead>
    <tr>
        <th rowspan="2">Method</th>
        <th rowspan="2">Type</th>
        <th rowspan="2">Count</th>
        <th rowspan="2">Differ</th>
        <th rowspan="2">Compare</th>
        <th colspan="2">Logical</th>
        <th rowspan="2">Overall</th>
    </tr>
    <tr>
        <th>Negate</th>
        <th>Universal</th>
    </tr>
    </thead>
    <tbody>
    <tr align="center">
        <td>Show-o</td>
        <td>Discrete Diff.</td>
        <td>0.70</td>
        <td>0.62</td>
        <td>0.71</td>
        <td>0.51</td>
        <td>0.65</td>
        <td>0.60</td>
    </tr>
    <tr align="center">
        <td>VILA-U</td>
        <td>Autoregressive</td>
        <td>0.70</td>
        <td>0.71</td>
        <td>0.74</td>
        <td>0.53</td>
        <td>0.66</td>
        <td>0.64</td>
    </tr>
    <tr align="center">
        <td>Liquid</td>
        <td>Autoregressive</td>
        <td>0.76</td>
        <td>0.73</td>
        <td>0.74</td>
        <td>0.46</td>
        <td>0.74</td>
        <td>0.65</td>
    </tr>
    <tr align="center">
        <th>UniTok</th>
        <th>Autoregressive</th>
        <th>0.76</th>
        <th>0.79</th>
        <th>0.74</th>
        <th>0.46</th>
        <th>0.73</th>
        <th>0.67</th>
    </tr>
    </tbody>
</table>

Please refer to [EVAL.md](https://github.com/FoundationVision/UniTok/blob/main/eval/EVAL.md) for more details.

### Evaluation

We also benchmark UniTok in terms of both understanding performance using the [LLaVA](https://github.com/haotian-liu/LLaVA) framework
and generation performance using the [LLamaGen](https://github.com/FoundationVision/LlamaGen) framework.
Please refer to [EVAL.md](https://github.com/FoundationVision/UniTok/blob/main/eval/EVAL.md) for more details.

## Acknowledgement
UniTok is built upon the awesome works
[VAR](https://github.com/FoundationVision/VAR),
[DataComp](https://github.com/mlfoundations/datacomp),
[Liquid](https://github.com/FoundationVision/Liquid/),
[LLaVA](https://github.com/haotian-liu/LLaVA/),
[LlamaGen](https://github.com/FoundationVision/LlamaGen/),
and [ViTamin](https://github.com/Beckschen/ViTamin).

## License

This project is licensed under the MIT License. See the [LICENSE](https://github.com/FoundationVision/UniTok/blob/main/LICENSE) file for details.

## Citation

If you find this project useful, please consider citing:

```bibtex
@article{unitok,
  title={UniTok: A Unified Tokenizer for Visual Generation and Understanding},
  author={Ma, Chuofan and Jiang, Yi and Wu, Junfeng and Yang, Jihan and Yu, Xin and Yuan, Zehuan and Peng, Bingyue and Qi, Xiaojuan},
  journal={arXiv preprint arXiv:2502.20321},
  year={2025}
}
```