Instructions to use UCSC-VLAA/openvision3-vit-base-patch2-32 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use UCSC-VLAA/openvision3-vit-base-patch2-32 with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:UCSC-VLAA/openvision3-vit-base-patch2-32') tokenizer = open_clip.get_tokenizer('hf-hub:UCSC-VLAA/openvision3-vit-base-patch2-32') - Notebooks
- Google Colab
- Kaggle
Commit History
-
Add model card for OpenVision 3
Verified signatureThis PR adds a model card for OpenVision 3, a family of unified visual encoders designed for both image understanding and generation. The card includes: - Metadata for `pipeline_tag`, `library_name` (`open_clip`), and `license`. - A brief description of the architecture and training objectives. - Links to the paper, project page, and GitHub repository. - A sample usage code snippet based on the official documentation. - Citation information.5970ea6
-
Copy from Letian2003/unified_vit_v66-vision-only
Verified signature7c5fbd3 -
initial commit
Verified signaturee076e9e