PRIME-RL

/

P1-30B-A3B

like 10

Follow

PRIME 113

This PR enhances the model card by:

• Adding the library_name: transformers to the metadata. This is supported by the Usage section's code snippet (from transformers import AutoModelForCausalLM, AutoTokenizer) and config.json ("transformers_version": "4.51.0"), enabling the automated "How to use" widget on the Hugging Face Hub.
• Improving the top section of the model card to include direct links to the paper (P1: Mastering Physics Olympiads with Reinforcement Learning) and the GitHub repository (https://github.com/PRIME-RL/P1), alongside the existing project page and leaderboard links, for better visibility and discoverability.
• Updating the url in the Citation section to point to the official Hugging Face paper page, ensuring the reference is accurate and directly accessible.