hustvl

/

vgt_internvl3_1_6B_pretrain

like 0

Follow

HUST Vision Lab 101

This PR adds a comprehensive model card for the VGT (Visual Generation Tuning) model, based on the official paper and GitHub repository.

The updates include:

• A clear description of the model and its capabilities.
• Links to the paper (Visual Generation Tuning) and the official GitHub repository (https://github.com/hustvl/VGT).
• The pipeline_tag: text-to-image metadata for improved discoverability on the Hugging Face Hub, reflecting the model's core functionality.
• Verbatim installation instructions and inference code snippets from the GitHub README to help users get started quickly, ensuring no code is made up.
• Integration of key highlights, a "What is VGT?" summary, a table of pretrained models, and the citation information.
• Visual examples from the GitHub repository.

Please review and merge this PR.