---
license: mit
---


# Model Card for ArborCLIP

<!-- Banner links -->
<div style="text-align:center;">
  <a href="https://baskargroup.github.io/Arboretum/" target="_blank">
    <img src="https://img.shields.io/badge/Project%20Page-Visit-blue" alt="Project Page" style="margin-right:10px;">
  </a>
  <a href="https://github.com/baskargroup/Arboretum" target="_blank">
    <img src="https://img.shields.io/badge/GitHub-Visit-lightgrey" alt="GitHub" style="margin-right:10px;">
  </a>
  <a href="https://pypi.org/project/arbor-process/" target="_blank">
    <img src="https://img.shields.io/badge/PyPI-arbor--process%200.1.0-orange" alt="PyPI arbor-process 0.1.0">
  </a>
</div>


ARBORCLIP is a new suite of vision-language foundation models for biodiversity. These models CLIP-style foundation models were trained on [ARBORETUM-40M](https://baskargroup.github.io/Arboretum/), which is a large-scale dataset of 40 million images of 100000+ species of plants and animals.


## Model detals
We released three trained models checkpoints in the [model_ckpt](#directory) directory. The models are trained on the Arboretum dataset.