OmniGen

by Beijing Academy of Artificial Intelligence

About the model:

Model designed to generate images from multimodal prompts.

Model type:

Image

Model performance class:

Full

Link to the model:

https://huggingface.co/Shitao/OmniGen-v1

Base models:

(undefined)

End model:

OmniGen-v1

End model license:

MIT License

About the organisation:

Chinese non-profit AI lab.

Link to the organisation:

https://www.baai.ac.cn/en/

Model release date:

October 2024

Availability

Base Model Data

Dataset made available, public datasets used to create it outlined in paper.

https://huggingface.co/collections/yzwang/x2i-dataset-674c66d1d700f7f816a9590d https://arxiv.org/pdf/2409.11340

End User Model Data

Base Model Weights

Weights made available on HuggingFace.

https://huggingface.co/Shitao/OmniGen-v1

End User Model Weights

Training Code

Only fine-tuning code made available on GitHub.

https://github.com/VectorSpaceLab/OmniGen

Documentation

Code Documentation

Fine-tuning code is sparsely documented.

https://github.com/VectorSpaceLab/OmniGen

Hardware Architecture

Training setup is sparsely documented.

https://arxiv.org/pdf/2409.11340

Preprint

Preprint made available on arXiv.

https://arxiv.org/pdf/2409.11340

Paper

Paper published in IEEE.

https://ieeexplore.ieee.org/document/11093711

Modelcard

Model card provides information about inference and fine-tuning, but not model architecture or training.

https://huggingface.co/Shitao/OmniGen-v1

Datasheet

Data thoroughly outlined.

https://huggingface.co/datasets/yzwang/X2I-subject-driven https://huggingface.co/datasets/yzwang/X2I-text-to-image https://huggingface.co/datasets/yzwang/X2I-mm-instruction https://huggingface.co/datasets/yzwang/X2I-in-context-learning https://huggingface.co/datasets/yzwang/X2I-computer-vision

Access

Licenses

MIT license, an OSI-approved license.

https://huggingface.co/Shitao/OmniGen-v1#license

Is this information not up to date?

Contribute here ->

Supported by the Centre for Language Studies and the Dutch Research Council. Website design & development © 2024 by BSTN. This version of the index generated 09 April 2026, website content last updated 11 March 2026.