marinero4972

/

Open-o3-Video

like 4

This PR significantly enhances the model card for the Open-o3 Video model by:

• Adding the pipeline_tag: video-text-to-text to accurately categorize the model's functionality (grounded video reasoning), improving discoverability on the Hub.
• Including library_name: transformers as the model uses Qwen2_5_VLForConditionalGeneration and Qwen2_5_VLProcessor components, indicating compatibility with the 🤗 Transformers library and enabling the automated "How to use" widget.
• Populating the content section with key information from the paper and the GitHub repository, including the abstract, model architecture details, and a comprehensive quick start guide covering environment setup, data preparation, training, and evaluation.
• Ensuring direct links to the Hugging Face paper page, project page, and GitHub repository are prominently displayed.

Please note that no sample usage code snippet has been added, as the provided GitHub README content only specifies an inference command (python ./inference_example.py) but does not include the actual code for direct integration as per documentation guidelines.

These updates will make the model more discoverable and easier to use for the community.