---
license: apache-2.0
pipeline_tag: text-to-image
library_name: llamagen
---

# Model Weights for ControlAR

<div style='display:flex; gap: 0.25rem; align-items: center'>
<a href="https://arxiv.org/abs/2410.02705"><img src="https://img.shields.io/badge/arXiv-Paper-red"></a>
<a href="https://github.com/hustvl/ControlAR"><img src="https://img.shields.io/badge/GitHub-Code-green"></a>
</div>

This model contains several checkpoints for [ControlAR](https://arxiv.org/abs/2410.02705)

| AR Model | Type | Control | Arbitrary-Resolution | Checkpoint |
| :--------| :--: | :-----: | :------------------: | :--------: |
| [LlamaGen-XL](https://github.com/FoundationVision/LlamaGen#-text-conditional-image-generation) | t2i | Canny Edge | ✅ | [ckpt](https://huggingface.co/wondervictor/ControlAR/blob/main/canny_MR.safetensors) |
| [LlamaGen-XL](https://github.com/FoundationVision/LlamaGen#-text-conditional-image-generation) | t2i | Depth | ✅ | [ckpt](https://huggingface.co/wondervictor/ControlAR/blob/main/depth_MR.safetensors) |
| [LlamaGen-XL](https://github.com/FoundationVision/LlamaGen#-text-conditional-image-generation) | t2i | HED Edge | ❌ | [ckpt](https://huggingface.co/wondervictor/ControlAR/blob/main/hed.safetensors) |
| [LlamaGen-XL](https://github.com/FoundationVision/LlamaGen#-text-conditional-image-generation) | t2i | Seg. Mask | ❌ | [ckpt](https://huggingface.co/wondervictor/ControlAR/blob/main/seg_cocostuff.safetensors) |


## Citation
If you find ControlAR is useful in your research or applications, please consider giving us a star 🌟 and citing it by the following BibTeX entry:

```bibtex
@article{li2024controlar,
      title={ControlAR: Controllable Image Generation with Autoregressive Models}, 
      author={Zongming Li, Tianheng Cheng, Shoufa Chen, Peize Sun, Haocheng Shen, Longjin Ran, Xiaoxin Chen, Wenyu Liu, Xinggang Wang},
      year={2024},
      eprint={2410.02705},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2410.02705}, 
}
```

## Acknowledgements
The development of ControlAR is based on [LlamaGen](https://github.com/FoundationVision/LlamaGen), [ControlNet](https://github.com/lllyasviel/ControlNet), [ControlNet++](https://github.com/liming-ai/ControlNet_Plus_Plus), and [AiM](https://github.com/hp-l33/AiM), and we sincerely thank the contributors for those great works!