Dongz

/

llama31-405b-quant

like 1

This PR significantly enhances the model card for llama31-405b-quant by providing more comprehensive and discoverable information.

Key updates include:

• Addition of license: apache-2.0, library_name: transformers, and pipeline_tag: text-generation to the metadata, improving categorization and clarity.
• Linking the model to its associated research paper: LLMC+: Benchmarking Vision-Language Model Compression with a Plug-and-play Toolkit.
• Including direct links to the project's official GitHub repository (https://github.com/ModelTC/llmc) and its documentation (https://llmc-en.readthedocs.io/en/latest/).
• Expanding the model description to provide context about its origin and the LightCompress (formerly LLMC) toolkit used for quantization.
• Providing a Python code snippet for loading and inferring with the model using the transformers library, including a chat template example suitable for Llama models.

These changes will make the model more accessible, informative, and usable for the Hugging Face community.