inclusionAI

/

Ring-1T-FP8

like 17

Follow

inclusionAI 1.19k

This PR significantly enhances the model card for the Ring-1T model by:

• Adding library_name: transformers to the metadata: This enables the automated "how to use" widget on the Hugging Face Hub, providing users with automated code snippets for easy integration with the transformers library.
• Aligning the main title of the model card with the official paper title: "Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model".
• Including a direct link to the Hugging Face paper page: Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model in the introductory section.
• Adding a prominent link to the GitHub repository: https://github.com/inclusionAI/Ring-V2 for quick access to the code.
• Integrating a transformers code snippet for quick model usage, as found in the original GitHub README, under the Quickstart section.
• Updating the SGLang and vLLM deployment sections with more comprehensive environment preparation and usage instructions from the GitHub repository.
• Adding the BibTeX citation for the paper.

These updates collectively improve the discoverability, usability, and completeness of the model card on the Hugging Face Hub.