dengcao

/

GLM-4.1V-9B-Thinking-GPTQ-Int4-Int8Mix

like 2

This PR significantly improves the model card for GLM-4.1V-9B-Thinking-GPTQ-Int4-Int8Mix by:

• Correcting the pipeline_tag to image-text-to-text. The model is a vision-language model, supporting multimodal inputs (images, videos, text) and generating text, which makes image-text-to-text the most accurate tag over the previous text-generation. This ensures the model appears in the correct categories on the Hub: https://huggingface.co/models?pipeline_tag=image-text-to-text.
• Adding library_name: transformers to the metadata. This enables the "Use in Transformers" button on the model page, providing direct code snippets for easy integration and use.
• Enhancing the model card content for better readability and completeness. Key sections such as "Model Introduction", "Benchmark Performance", "Model Inference", "Model Fine-tuning", and "Model License" have been updated and provided in English, using information directly from the official GitHub repository README, making the information more accessible to a global audience.
• Adding direct links to the paper (using the existing Arxiv link), the official GitHub repository, and relevant Hugging Face/ModelScope demos at the top of the model card.

These changes will greatly improve the discoverability, understanding, and usability of the model on the Hugging Face Hub.