Image Feature Extraction
MLX
Safetensors
vision-transformer
self-supervised-learning
dense-prediction
Instructions to use mnmly/lingbot-vision-vit-large-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mnmly/lingbot-vision-vit-large-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir lingbot-vision-vit-large-mlx mnmly/lingbot-vision-vit-large-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
| { | |
| "arch": "vit_large", | |
| "patchSize": 16, | |
| "inChannels": 3, | |
| "embedDim": 1024, | |
| "depth": 24, | |
| "numHeads": 16, | |
| "ffnRatio": 4.0, | |
| "imgSize": 512, | |
| "qkvBias": true, | |
| "projBias": true, | |
| "ffnBias": true, | |
| "layerscaleInit": 1e-05, | |
| "nStorageTokens": 4, | |
| "normLayer": "layernormbf16", | |
| "normEps": 1e-05, | |
| "ffnLayer": "mlp", | |
| "ropeBase": 100.0, | |
| "maskKBias": true | |
| } |