kijai/ComfyUI-LivePortraitKJ

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2200 stars · 176 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

ComfyUI custom nodes that integrate the LivePortrait face animation and retargeting model, supporting image-to-video, video-to-video, and near-realtime webcam driving. It offers alternative face detectors (MediaPipe, face-alignment) and ships models as safetensors with automatic download.

Use cases

• animate a portrait photo with a driving video
• retarget facial expressions from one video to another
• run realtime face animation from a webcam feed
• build ComfyUI workflows for talking-head video generation
• swap face detectors for better small-face detection
• avoid non-commercial Insightface license by using MediaPipe

When to choose

• you already use ComfyUI and want LivePortrait as nodes
• you need realtime or near-realtime portrait animation pipelines
• you need MIT/Apache-licensed face detection via MediaPipe
• you want vid2vid retargeting with smoothing

When to avoid

• you need a standalone CLI or app outside ComfyUI
• you require commercial use with Insightface detection
• you need production-stable workflows across version upgrades (breaking changes occurred)

Facets

plugin · maturity active

video-processing image-processing machine-learning computer-vision plugin-system image-processing artificial-intelligence computer-vision media python cross-platform comfyui liveportrait face-animation face-retargeting talking-head webcam safetensors custom-nodes video gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="kijai/ComfyUI-LivePortraitKJ")