---
license: mit
tags:
- mlx
- liveportrait
- image-to-video
- apple-silicon
base_model:
- KlingTeam/LivePortrait
- warmshao/FasterLivePortrait
- jdh-algo/JoyVASA
- TencentGameMate/chinese-hubert-base
---

# FasterLivePortrait-MLX Weights

Converted MLX `.npz` runtime weights for
[FasterLivePortrait-MLX](https://github.com/ivanfioravanti/fasterliveportrait-mlx).

These files are converted from permissively licensed source checkpoints:

- [KlingTeam/LivePortrait](https://huggingface.co/KlingTeam/LivePortrait), MIT
- [warmshao/FasterLivePortrait](https://huggingface.co/warmshao/FasterLivePortrait), MIT
- [jdh-algo/JoyVASA](https://huggingface.co/jdh-algo/JoyVASA), MIT
- [TencentGameMate/chinese-hubert-base](https://huggingface.co/TencentGameMate/chinese-hubert-base), MIT

## Included

- Human LivePortrait core MLX weights
- Human landmark MLX weights
- Human stitching / eye / lip retargeting MLX weights
- Animal LivePortrait v1.1 core MLX weights
- JoyVASA MLX audio-to-motion weights for the configured Chinese HuBERT path
- JoyVASA motion template runtime asset

## Not Included

This repository intentionally does **not** include XPose. XPose is used only for
animal landmark detection in FasterLivePortrait-MLX, and its upstream license is
restricted to non-commercial research use.

This repository also does **not** include the original JoyVASA PyTorch
checkpoint or the original Transformers HuBERT directory. Those are conversion
inputs only.

This repository also does not include the MediaPipe Face Landmarker task model;
download it from Google's MediaPipe model URL as documented in the project
README.

## Use

```bash
uv run python scripts/download_mlx_weights.py --repo-id ivanfioravanti/FasterLivePortrait-MLX-weights
```

## Conversion

The weights were produced with:

```bash
uv run --group convert python scripts/export_mlx_weights.py --include-animal
```

JoyVASA weights were produced with:

```bash
uv run --group convert python scripts/export_mlx_weights.py --include-joyvasa
```

Converted tensors are derivative model weights and inherit the obligations of
the original model licenses.