ymccccc

/

GS-Reasoner-Diverse

like 0

This PR significantly enhances the model card for GS-Reasoner by adding essential metadata and structured content.

Changes include:

• Adding library_name: transformers to enable the automated "how to use" widget, based on evidence from config.json and tokenizer_config.json.
• Specifying pipeline_tag: image-text-to-text to improve discoverability on the Hub, reflecting its function of processing visual (3D context) and text input to generate text output.
• Including additional descriptive tags like 3d-vision, visual-grounding, and spatial-reasoning.
• Providing links to the official paper, project page, and GitHub repository.
• Incorporating the full paper abstract for detailed context.
• Adding the installation/setup commands from the GitHub README.
• Detailing the available model weights.

Please review and merge this PR.