How to use xiaomoguhzz/VisionEncoder with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("xiaomoguhzz/VisionEncoder", dtype="auto")
