xiabs

/

DreamOmni2

like 85

This PR significantly enhances the model card for DreamOmni2 by adding crucial metadata and comprehensive content.

Specifically, it adds:

• pipeline_tag: text-to-image, enabling the model to be easily discoverable via Hugging Face's pipeline filtering.
• library_name: transformers, to provide an automated code snippet for seamless integration with the Hugging Face transformers library.
• A direct link to the paper: DreamOmni2: Multimodal Instruction-based Editing and Generation.
• A link to the project page: https://pbihao.github.io/projects/DreamOmni2/index.html.
• A link to the GitHub repository: https://github.com/dvlab-research/DreamOmni2.
• Detailed "Quick Start" instructions and Python code snippets for both multimodal instruction-based editing and generation, extracted directly from the GitHub README.
• Web demo instructions and a BibTeX citation entry.

These additions will greatly enhance the model's discoverability, usability, and overall documentation for the Hugging Face community.

Please review and merge this PR if everything looks good.