RISys-Lab
/

ReasonCLIP-L14-336-S2

Zero-Shot Image Classification

Model card Files Files and versions

ReasonCLIP-L14-336-S2 / README.md

fesvhtr's picture

docs: update usage model id namespace

1832b32 verified 4 days ago

|

History Blame Contribute Delete

1.09 kB

	---
	library_name: transformers
	license: cc-by-nc-sa-4.0
	tags: []
	---

	## Model Details

	- Model: ReasonCLIP-L14-336-S2
	- Base model: [openai/clip-vit-large-patch14-336](https://huggingface.co/openai/clip-vit-large-patch14-336)
	- Architecture: CLIP ViT-L/14
	- Image resolution: 336
	- Training stage: Stage 2
	- Training data: [ReasonLite-42M](https://huggingface.co/datasets/RISys-Lab/ReasonCLIPLite-42M) and [ReasonPro-16M](https://huggingface.co/datasets/RISys-Lab/ReasonCLIPPro-16M)

	## Method

	![Method overview](https://raw.githubusercontent.com/RISys-Lab/ReasonCLIP/main/doc/method.png)

	## Resources

	- GitHub: [RISys-Lab/ReasonCLIP](https://github.com/RISys-Lab/ReasonCLIP)
	- Paper: [arXiv:2606.26794](https://arxiv.org/abs/2606.26794)

	## Usage

	```python
	from transformers import CLIPModel, CLIPProcessor

	model_id = "RISys-Lab/ReasonCLIP-L14-336-S2"
	model = CLIPModel.from_pretrained(model_id)
	processor = CLIPProcessor.from_pretrained(model_id)
	```

	For the full checkpoint list, see the [ReasonCLIP model card](https://github.com/RISys-Lab/ReasonCLIP/blob/main/doc/model_card.md).