Zero-Shot Classification
MLX
Safetensors
qwen3_5
clef
cloudflare
systemone
structured-output
classification
multimodal
custom-code
decision-model
Eval Results
4-bit precision
Instructions to use mlx-community/clef-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/clef-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download mlx-community/clef-4bit --local-dir clef-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
clef_mlx.py: default --model works inside Hub cache snapshots (symlinked files)
Browse files- clef_mlx.py +2 -1
clef_mlx.py
CHANGED
|
@@ -585,7 +585,8 @@ def load(path: str | Path, head_dtype=mx.bfloat16, backend: str = "auto") -> Cle
|
|
| 585 |
|
| 586 |
def _default_model() -> str | None:
|
| 587 |
"""When run from inside a downloaded repo, default to that repo."""
|
| 588 |
-
|
|
|
|
| 589 |
return str(here) if (here / "joint_head.safetensors").exists() else None
|
| 590 |
|
| 591 |
|
|
|
|
| 585 |
|
| 586 |
def _default_model() -> str | None:
|
| 587 |
"""When run from inside a downloaded repo, default to that repo."""
|
| 588 |
+
# absolute(), not resolve(): Hub cache snapshots are symlinks into a blob store.
|
| 589 |
+
here = Path(__file__).absolute().parent
|
| 590 |
return str(here) if (here / "joint_head.safetensors").exists() else None
|
| 591 |
|
| 592 |
|