MMR1

/

MMR1-32B-SFT

like 0

Follow

MMR1 12

This PR significantly enhances the model card for MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources.

Key improvements include:

• Adding license: apache-2.0 based on the GitHub README.
• Specifying library_name: transformers as evidenced by the model's configuration files.
• Setting pipeline_tag: image-text-to-text for better discoverability of this multimodal reasoning model.
• Including relevant tags: multimodal-llm, reasoning, and math.
• Adding datasets metadata for MMR1/MMR1-SFT and MMR1/MMR1-RL as listed in the project's open resources.
• Integrating a comprehensive description of the model, its methodology, open resources, evaluation, and qualitative demo directly from the project's GitHub README.
• Ensuring proper links to the paper and the GitHub repository.

No sample usage is included as a verifiable transformers-compatible code snippet was not found in the provided GitHub README content, as per instructions.