Emu

<div align='center'> <h1>Emu: An Open Multimodal Generalist</h1h1> <h3><a href="https://arxiv.org/abs/2307.05222">Generative Pretraining in Multimodality</a></h3>

Quan Sun^1*, Qiying Yu^2,1*, Yufeng Cui^1*, Fan Zhang^1*, Xiaosong Zhang^1*, Yueze Wang¹, Hongcheng Gao¹, Jingjing Liu², Tiejun Huang^1,3, Xinlong Wang¹

¹ BAAI, ² THU, ³ PKU <br>^* Equal Contribution

| Paper | Demo(tmp) |

</div>

Emu is a Large Multimodal Model (LMM) trained with a unified autoregressive objective, i.e., predict-the-next-element, including both visual embeddings and textual tokens. Trained under this objective, Emu can serve as a generalist interface for diverse multimodal tasks, such as image captioning, image/video question answering, and text-to-image generation, together with new abilities like in-context text and image generation, and image blending.

Setup

Clone the github repository and install required packages:

git clone https://github.com/baaivision/Emu
cd Emu

pip install -r requirements.txt

Model Weights

We release the pretrained and instruction-tuned weights of Emu. Our weights are subject to LLaMA's license.

Model Usage

At present, we provide inference code for image captioning and visual question answering:

python emu_inference.py --instruct --ckpt-path $Instruct_CKPT_PATH

Acknowledgement

We thank the great work from LLaMA, BLIP-2, Stable Diffusion, and FastChat.

Citation

If you find Emu useful for your your research and applications, please consider citing:

@article{Emu,
  title={Generative Pretraining in Multimodality},
  author={Sun, Quan and Yu, Qiying and Cui, Yufeng and Zhang, Fan and Zhang, Xiaosong and Wang, Yueze and Gao, Hongcheng and Liu, Jingjing and Huang, Tiejun and Wang, Xinlong},
  publisher={arXiv:2307.05222},
  year={2023},
}