ModelPrésentéExterne

Fastvit Mci3.apple Mclip2 Dfndr2b

A MobileCLIP v2 (image encoder only) for timm. Equivalent to image tower from

Sourcehuggingface.co/timm/fastvit_mci3.apple_mclip2_dfndr2b/tree/5932820f90e3d903380216e9563e102c909a5bde

TypeOtherVersionRevision 5932820CClasse

Source du modèle

Type: Other
Taille du fichier: 479 MB
Version: vRevision 5932820
Mis à jour: 17 sept. 2026
Téléchargements: 79
Source: Hugging Face

Description de la source

Model card for fastvit_mci3.apple_mclip2_dfndr2b

A MobileCLIP v2 (image encoder only) for timm. Equivalent to image tower from https://huggingface.co/timm/MobileCLIP2-S3-OpenCLIP.

Model Details

Dataset: DFNDR-2B
Papers:
- MobileCLIP2: Improving Multi-Modal Reinforced Training: https://arxiv.org/abs/2508.20691

Citation

@article{faghri2025mobileclip2,
          title={MobileCLIP2: Improving Multi-Modal Reinforced Training},
          author={Faghri, Fartash and Vasu, Pavan Kumar Anasosalu and Koc, Cem and Shankar, Vaishaal and Toshev, Alexander and Tuzel, Oncel and Pouransari, Hadi},
          journal={arXiv preprint arXiv:2508.20691},
          year={2025}
        }

Sources

1 source

Hugging FacePrincipalVersion 5932820f90e3d903380216e9563e102c909a5bde

huggingface.co/timm/fastvit_mci3.apple_mclip2_dfndr2b/tree/5932820f90e3d903380216e9563e102c909a5bde

Vérifié 2 sept.

Download files

Artefacts du modèle

2 artefacts

model.safetensors

safetensors · 479 MB · SHA-256 5b96d3e2dcb2…9149 · Hugging Face

Download file

pytorch_model.bin

bin · 479 MB · SHA-256 d47c88e543eb…047a · Hugging Face

Download file

Extraits de sources

2 extraits

Readme2 sept.huggingface.co/timm/fastvit_mci3.apple_mclip2_dfndr2b

--- tags: - timm - transformers - image-feature-extraction - mobileclip - mobileclip2 library_name: timm license: apple-amlr datasets: - dfndr-2b --- # Model card for fastvit_mci3.apple_mclip2_dfndr2b A MobileCLIP v2 (image encoder only) for `timm`. Equivalent to image tower from https://huggingface.co/timm/MobileCLIP2-S3-OpenCLIP. ## Model Details - **Dataset:** DFNDR-2B - **Papers:** - MobileCLIP2: Improving Multi-Modal Reinforced Training: https://arxiv.org/abs/2508.20691 ## Citation ```bibtex @article{faghri2025mobileclip2, title={MobileCLIP2: Improving Multi-Modal Reinforced Training}, author={Faghri, Fartash and Vasu, Pavan Kumar Anasosalu and Koc, Cem and Shankar, Vaishaal and Toshev, Alexander and Tuzel, Oncel and Pouransari, Hadi}, journal={arXiv preprint arXiv:2508.20691}, year={2025} } ```

Note2 sept.huggingface.co/timm/fastvit_mci3.apple_mclip2_dfndr2b

Source context: 79 downloads · 0 likes · Pipeline image-feature-extraction · Library timm · Repo timm/fastvit_mci3.apple_mclip2_dfndr2b