Improve model card: Update pipeline_tag, add library_name, and include NABLA paper details
#10
by
nielsr
HF Staff
- opened
This PR enhances the model card for the Wan2.1 model by:
- Updating the
pipeline_tagfromimage-to-videotoany-to-anyto accurately reflect the model's diverse generative capabilities, which include Text-to-Video, Image-to-Video, Video Editing, Text-to-Image, and Video-to-Audio. This improves discoverability on the Hub. - Adding
library_name: diffusersto the metadata, as the model explicitly provides usage examples with the Diffusers library, enabling the "Use in Diffusers" widget. - Including a new section to introduce $\nabla$NABLA: Neighborhood Adaptive Block-Level Attention (https://huggingface.co/papers/2507.13546). This paper describes a core attention mechanism that contributes to the efficiency and performance of video diffusion transformers like Wan2.1, providing crucial context for the model.