ComfyUI-AdvancedLivePortrait-Universal

ComfyUI-AdvancedLivePortrait (Universal Edition)

⚠️ Maintained Fork Notice: This is a maintained community edition of the original PowerHouseMan/ComfyUI-AdvancedLivePortrait. This fork brings native multi-device acceleration (Intel XPU, Apple Silicon MPS, AMD ROCm, NVIDIA CUDA, and CPU) via ComfyUI model management.

🌐 Universal Device & Hardware Support

Unlike the original CUDA-only implementation, this fork integrates seamlessly with ComfyUI's native device management:

• Intel Arc & Data Center GPUs (XPU) — Full acceleration via PyTorch XPU backend
• NVIDIA GPUs (CUDA) — Native CUDA acceleration
• Apple Silicon (MPS) — Metal Performance Shaders acceleration on Mac (M1/M2/M3/M4)
• AMD GPUs (ROCm / DirectML) — Supported through ComfyUI device mapping
• CPU Fallback — Automatically supported if ComfyUI is launched in CPU mode (--cpu)

Features

• ExpressionEditor: Interactive slider-based expression adjustments (smile, wink, eyebrow, gaze, mouth shapes, head pose) with real-time UI preview.
• Animation Chaining: Multi-keyframe facial expression sequences with automated smoothing and interpolation.
• Expression Presets: Save and load lightweight reusable .exp expression files (SaveExpData, LoadExpData).

---

Introduction

AdvancedLivePortrait is faster and has real-time preview

https://github.com/user-attachments/assets/90b78639-6477-48af-ba49-7945488df581

• Edit facial expressions in photos.
• Insert facial expressions into videos.
• Create animations using multiple facial expressions (with or without driving video).
• Extract facial expressions from sample photos.
• Track faces across video frames.

Installation

Clone this repository into your ComfyUI/custom_nodes/ directory:

cd custom_nodes
git clone https://github.com/cyberterm/ComfyUI-AdvancedLivePortrait-Universal.git

Usage

Workflows and sample data are located in custom_nodes/ComfyUI-AdvancedLivePortrait-Universal/sample/workflows/

---

You can add expressions to the video. See Advanced_Inserting_expressions_into_vid.json or other sample workflows.

Describes the 'command' in the workflow:

[Motion index] = [Changing frame length] : [Length of frames waiting for next motion]

• Motion index 0 is the original source image.
• They are numbered in the order they lead to the motion_link.
• Linking the driving video to src_images will add facial expressions to the driving video.

---

You can save and load expressions with the Load Exp Data and Save Exp Data nodes.

Expression data is saved to \ComfyUI\output\exp_data\.

---

Thanks

• Original author's link : https://liveportrait.github.io/
• Upstream ComfyUI implementation: https://github.com/PowerHouseMan/ComfyUI-AdvancedLivePortrait
• This project uses models converted by kijai: https://github.com/kijai/ComfyUI-LivePortraitKJ