showlab

/

show-o-512x512-wo-llava-tuning

like 1

Follow

Show Lab 216

This PR significantly improves the model card for showlab/show-o2 by replacing the generic template content with detailed information extracted from the paper abstract and the project's GitHub repository.

Key updates include:

• A comprehensive model description and key capabilities, leveraging the paper's abstract and GitHub overview.
• Updated and added external links, including the specific GitHub repository path (show-o/tree/main/show-o2) and a direct link to the Hugging Face Space demo.
• Integration of the detailed "News" section from the GitHub README, providing a chronological overview of project updates and features.
• Inclusion of visual examples (GIFs and images) directly from the GitHub repository to showcase the model's performance.
• A detailed "How to Get Started" section with practical code snippets for inference (Multimodal Understanding, Text-to-Image Generation).
• An overview of the multi-stage training pipeline.
• Full BibTeX citations for both the Show-o2 and the original Show-o papers.
• Populated Developed by and Language(s) fields.

This update transforms the model card into a much more informative and user-friendly resource for anyone interested in Show-o2.