Instructions to use MedMLX/synthseg-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use MedMLX/synthseg-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download MedMLX/synthseg-mlx --local-dir synthseg-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Install the published PyPI release
Browse files
README.md
CHANGED
|
@@ -28,11 +28,10 @@ they have not been retrained. Convolution kernels use MLX's
|
|
| 28 |
| `synthseg_qc_2.0.safetensors` | Quality control |
|
| 29 |
|
| 30 |
Every variant uses the parcellation and QC files, even when parcellation is
|
| 31 |
-
disabled. On Apple silicon with MLX and Metal, install the
|
| 32 |
-
runtime from its pinned source revision:
|
| 33 |
|
| 34 |
```bash
|
| 35 |
-
uv pip install "synthseg-mlx
|
| 36 |
```
|
| 37 |
|
| 38 |
Then download the weights and run:
|
|
|
|
| 28 |
| `synthseg_qc_2.0.safetensors` | Quality control |
|
| 29 |
|
| 30 |
Every variant uses the parcellation and QC files, even when parcellation is
|
| 31 |
+
disabled. On Apple silicon with MLX and Metal, install the runtime:
|
|
|
|
| 32 |
|
| 33 |
```bash
|
| 34 |
+
uv pip install "synthseg-mlx>=0.3.1" huggingface-hub
|
| 35 |
```
|
| 36 |
|
| 37 |
Then download the weights and run:
|