---
name: DFN-public-SAFETENSORS
base_model: openai/clip-vit-base-patch32
license: apple-amlr
pipeline_tag: zero-shot-image-classification
tags: 
- clip
- Apple
- OpenAI
size:
- 605159932
- 605MB
tasks:
 - contrastive image-text
 - vision
language: en
papers:
- https://arxiv.org/abs/2309.17425
datasets:
- CommonPool-12.8B
license_link: LICENSE
---
> [!IMPORTANT]
> Original Model Link : [https://huggingface.co/apple/DFN-public](https://huggingface.co/apple/DFN-public)
> 

```
name: DFN-public-SAFETENSORS
base_model: openai/clip-vit-base-patch32
license: apple-amlr
pipeline_tag: zero-shot-image-classification
tags: 
- clip
- Apple
- OpenAI
size:
- 605159932
- 605MB
tasks:
 - contrastive image-text
 - vision
language: en
papers:
- https://arxiv.org/abs/2309.17425
datasets:
- Conceptual Captions 12M
- Conceptual Captions 3M
- Shutterstock 15M.
license_link: LICENSE
```


# DFN-public-SAFETENSORS
A Drop-in replacement for OpenCLIP trained on Data Filtering Network from CC12M, CC3M, and SS15M