Add comprehensive model card for MM-HELIX-7B-Thinking

#1
by nielsr HF Staff - opened

This PR adds a comprehensive model card for the MM-HELIX-7B-Thinking model.

It includes:

  • The license: cc-by-nc-4.0 tag.
  • The library_name: transformers tag, enabling an automated "How to use" widget on the Hub.
  • The pipeline_tag: video-text-to-text to improve discoverability, reflecting its multimodal video processing capabilities.
  • Additional tags (multimodal, video, reasoning, qwen) for better categorization.
  • Links to the official paper, project page, GitHub repository, and associated Hugging Face datasets.
  • A detailed description of the model, its methodology (AHPO, SERG), the MM-HELIX benchmark, and evaluation results, largely adapted from the project's GitHub README.

Please review these additions to ensure the model card accurately represents the MM-HELIX project and assists users in understanding and utilizing the model.

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment