lazyslide.models.multimodal.MUSK

class MUSK(model_path=None, token=None, multiscale_augmentation=False, with_head=True)

Bases: ImageTextModel

🤗Hugging Face GitHub Paper Params: 675.2M 1024 features FLOPs: 382.13G CC-BY-NC-ND-4.0 [Xiang et al., 2025] A Vision-Language Foundation Model for Precision Oncology

encode_image(image)

This should return the image feature before normalize.