Instructions to use apocryphx/gemma-4-31b-it-qat-q4-g32-apml with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use apocryphx/gemma-4-31b-it-qat-q4-g32-apml with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir gemma-4-31b-it-qat-q4-g32-apml apocryphx/gemma-4-31b-it-qat-q4-g32-apml
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download manifest.json from apocryphx/gemma-4-31b-it-qat-q4-g32-apml: direct link, hf CLI and curl.
- Browser
- Download file 825 Bytes
-
https://huggingface.co/apocryphx/gemma-4-31b-it-qat-q4-g32-apml/resolve/main/manifest.json
- Command line
-
hf download hf://apocryphx/gemma-4-31b-it-qat-q4-g32-apml/manifest.json
-
curl -L -o manifest.json https://huggingface.co/apocryphx/gemma-4-31b-it-qat-q4-g32-apml/resolve/main/manifest.json
825 Bytes
| { | |
| "config" : "config.json", | |
| "source" : { | |
| "revision" : "", | |
| "model_id" : "\/Users\/apocryphx\/.cache\/huggingface\/hub\/models--google--gemma-4-31B-it-qat-q4_0-unquantized\/snapshots\/4f926903562062220b3e54c1385c5ef2cd40bfd1" | |
| }, | |
| "variants" : [ | |
| { | |
| "path" : "weights\/mlx-q4", | |
| "id" : "mlx-q4", | |
| "quantization" : { | |
| "embed_bits" : 8, | |
| "bits" : 4, | |
| "scheme" : "mlx-affine", | |
| "group_size" : 32 | |
| }, | |
| "precision" : "q4", | |
| "files" : [ | |
| "model.safetensors" | |
| ], | |
| "runtime" : "mlx" | |
| } | |
| ], | |
| "tokenizer" : { | |
| "file" : "tokenizer.json", | |
| "kind" : "huggingface-tokenizers" | |
| }, | |
| "chat_template" : "chat_template.jinja", | |
| "architecture" : "gemma4", | |
| "kind" : "apertura-model", | |
| "format_version" : 1, | |
| "default_variant" : "mlx-q4" | |
| } |