ViT pretrained models avalible for downloading

[EdyQiu]

Hi, I found some pre-trained ViT models on the official web, https://mmpretrain.readthedocs.io/en/latest/papers/vision_transformer.html,
These models have top performance, but they are too expensive for me, unfortunately. The reviewers suggested I provide results using ViT models in the next submission, I am wondering whether the most "default" and "cheap" pre-trained ViT model could be provided (just like ResNet 50 for CNN). I think the corresponding config might be "vit-base-p16_64xb64_in1k.py" (Sorry I am not that familiar with ViT settings).