Feature Extraction
MLX
Safetensors
muse_glimmer
mlx-vlm
metaencoder
multimodal
embeddings
quantized
8-bit precision
Instructions to use edgefloor/meta-encoder-mlx-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use edgefloor/meta-encoder-mlx-8bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download edgefloor/meta-encoder-mlx-8bit --local-dir meta-encoder-mlx-8bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download conversion.json from edgefloor/meta-encoder-mlx-8bit: direct link, hf CLI and curl.
- Browser
- Download file 285 Bytes
-
https://huggingface.co/edgefloor/meta-encoder-mlx-8bit/resolve/main/conversion.json
- Command line
-
hf download hf://edgefloor/meta-encoder-mlx-8bit/conversion.json
-
curl -L -o conversion.json https://huggingface.co/edgefloor/meta-encoder-mlx-8bit/resolve/main/conversion.json
285 Bytes
| { | |
| "source": "facebook/meta-encoder", | |
| "revision": "3d0df704aa5bf66d9c49da1393181f33e893e5f7", | |
| "bits": 8, | |
| "group_size": 64, | |
| "mode": "affine", | |
| "vision": "unchanged BF16", | |
| "source_tensors": 1436, | |
| "quantized_modules": 418, | |
| "weight_bytes": 33440700416, | |
| "mlx": "0.32.3" | |
| } | |