This model is the stage 2 checkpoint of one of the thirteen settings, CLIP+DINOv2@336, used in the Law of Vision Representation in MLLMs.
- Downloads last month
- 22
Inference API (serverless) does not yet support transformers models for this pipeline type.