Instructions to use Avdpro/MLX-LivePortrait with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Avdpro/MLX-LivePortrait with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download Avdpro/MLX-LivePortrait --local-dir MLX-LivePortrait
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
|
Download NOTICE.md from Avdpro/MLX-LivePortrait: direct link, hf CLI and curl.
- Browser
- Download file 584 Bytes
-
https://huggingface.co/Avdpro/MLX-LivePortrait/resolve/main/NOTICE.md
- Command line
-
hf download hf://Avdpro/MLX-LivePortrait/NOTICE.md
-
curl -L -o NOTICE.md https://huggingface.co/Avdpro/MLX-LivePortrait/resolve/main/NOTICE.md
584 Bytes
MLX LivePortrait checkpoint notices
The seven weights/*.npz files are deterministic MLX conversions of the
official human LivePortrait checkpoint at KlingTeam/LivePortrait revision
82a4fa6735ca58432b6ce39301b4b9ee066dea47. Conversion changes tensor layout
for MLX execution and folds inference spectral normalization; it does not train
or alter the learned model.
The detector bundle contains face_detection_yunet_2023mar.onnx converted to
the oMLX graph format. YuNet is licensed under the MIT License included as
YUNET_LICENSE.
No InsightFace checkpoint is included.