Web-SSL DINO ViT-5B: 2B MetaCLIP data, 224 Resolution
A 5 billion parameter Vision Transformer (ViT) trained with DINOv2 self-supervised learning on web-scale image data without language supervision. Introduced in "Scaling Language-Free Visual Representation Learning" (Fan et al., 2025).
Model Details
- Architecture: ViT (3584 width, 32 depth, 28 heads)
- Parameters: 5B
- Resolution: 224×224 pixels
- Training: Self-supervised Web-DINO on 2B image samples from MetaCLIP web data
Model Descriptions
Web-SSL DINO 5B is a 5 billion parameter Vision Transformer model trained using self-supervised learning on 2 billion web images without language supervision. This model demonstrates that pure visual learning, when scaled appropriately, can match or exceed the performance of language-supervised models like CLIP across various vision tasks. It performs well on both traditional vision benchmarks and multimodal tasks including visual question answering and OCR & chart understanding.
Usage
from transformers import AutoImageProcessor, Dinov2Model
import torch
from PIL import Image
processor = AutoImageProcessor.from_pretrained('facebook/webssl-dino5b-full2b-224')
model = Dinov2Model.from_pretrained('facebook/webssl-dino5b-full2b-224')
# Process an image
image = Image.open('path/to/image.jpg')
inputs = processor(images=image, return_tensors="pt")
with torch.no_grad():
outputs = model(**inputs)
cls_features = outputs.last_hidden_state[:, 0] # CLS token features
patch_features = outputs.last_hidden_state[:, 1:] # patch-wise token features
Citation
@article{fan2025scaling,
title={Scaling Language-Free Visual Representation Learning},
author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
year={2025},
eprint={2504.01017},
archivePrefix={arXiv},
primaryClass={cs.CV}
}