Edit model card

This model is the stage 2 checkpoint of one of the thirteen settings, CLIP+DINOv2@224, used in the Law of Vision Representation in MLLMs.

Downloads last month
10
Safetensors
Model size
7.37B params
Tensor type
BF16
·
Inference API
Inference API (serverless) does not yet support transformers models for this pipeline type.