Rad Dino is a model hosted on Hugging Face for image feature extraction. Developed by Microsoft and based on the DINOv2 self-supervised vision approach, it generates embeddings from input images and is provided under the name microsoft/rad-dino.
The model can be loaded and run through the Transformers library. Users instantiate it with a pipeline for image-feature-extraction or directly via AutoImageProcessor and AutoModel classes, with support for device mapping. It is distributed in Safetensors format along with associated configuration files and weights intended for fine-tuning. The repository includes references to the underlying DINOv2 research papers and a model card that outlines usage, training details, and evaluation information.
It is delivered as an open model card on the Hugging Face platform, where it can be accessed for local inference, notebooks on Google Colab or Kaggle, or deployment through Hugging Face inference providers. The page indicates compatibility with the Transformers library and provides example code for both high-level pipelines and direct model loading.
No pricing information is stated; the model is available for download and use following the standard practices of public Hugging Face repositories.
Rad Dino is an Other AI product. Insufficient high-quality visual representations for radiology and medical imaging analysis. It is built as an open-source project for developers. Rad Dino is open source under the Apache-2.0 license. Rad Dino is available on the web, the command line, and API.
Behind Rad Dino is Microsoft, based in the United States, and the product first shipped in 2023. Development happens publicly on GitHub with 13.1k stars and 1 commits in the last 90 days. Key capabilities include image feature extraction, medical imaging, and transformers support.
Latest indexed changes and source events
microsoft/rad-dino verified by the PulseGate indexer
Other apps tracked under the same category.
