Appofficial publisher

LivePortrait

App : A product you use directly

Animates a portrait from a reference video, locally.

The best free tool to bring a photo to life.

Official site GitHub · 19,309 ★

Overall score

7.3 /10

Security: one caveat, see the card

Utility

8.5

Reliability

9.0

Security

One caveat

Ease

Technical · Technical

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Animating a portrait photo or video from a reference video has long required complex or expensive tools. LivePortrait offers a free, open-source solution that runs entirely on your own machine, without sending your data to a remote server.

In practice, you provide a source image (or video) of a face and a driving video containing the movements to reproduce. LivePortrait transfers the expressions, head movements and eye blinks from the reference video onto the source portrait. The result is exported as an MP4 file. The tool also includes an animals mode for animating portraits of cats and dogs.

LivePortrait is aimed at content creators, developers and curious users who want to experiment with portrait animation without a subscription or mandatory internet connection. A Gradio interface is available for those who prefer to avoid the command line.

However, installation requires technical skills: you need to manage a Python environment, install dependencies and, on Windows, download a separate installer. On macOS with Apple Silicon, processing can be around twenty times slower than on a recent NVIDIA GPU. Animals mode is not available on macOS.

What it offers

·Portrait animation from a driving video: expressions and head movements are transferred onto the source image.
·Image-driven mode: a single image can be used as a reference instead of a video.
·Portrait video editing (video-to-video): an existing video can be used as the source to animate.
·Animals mode: animates portraits of cats and dogs (available on Linux and Windows with NVIDIA GPU only).
·Regional control and precise portrait editing: allows adjustment of specific facial areas via the Gradio interface.
·Motion template generation (.pkl files): speeds up subsequent inferences and protects privacy by avoiding sharing the original video.
·Driving video auto-cropping: the --flag_crop_driving_video option automatically adjusts the framing.
·Built-in Gradio interface: provides a local graphical interface to use the tool without the command line.

Our tips

Crop your driving video to a 1:1 aspect ratio (e.g. 512x512 pixels) and focus on the face area for better results; alternatively, enable the --flag_crop_driving_video option.
Make sure the first frame of the driving video shows a frontal face with a neutral expression: this noticeably improves animation quality.
Minimise shoulder movement in the driving video, as the tool is optimised for head movements and facial expressions.
Use motion template files (.pkl) to re-run an animation on a new source image without reprocessing the driving video, which speeds up your workflow.
On macOS with Apple Silicon, add the environment variable PYTORCH_ENABLE_MPS_FALLBACK=1 before the inference command to avoid compatibility errors.

Watch out

On macOS with Apple Silicon, processing can be around twenty times slower than on a recent NVIDIA GPU, making it impractical for longer videos.
Animals mode is only available on Linux and Windows with an NVIDIA GPU; it is not supported on macOS.

In practice

Price: Free
Commercial use: Commercial use: yes
Available on: windows, linux, macos
Licence: MIT
Runs locally: Yes, 8 GB VRAM recommended
FR interface: No

Security and data

Rating: Security: one caveat, see the card
Your data: Data stays local
Maintenance: Active

Links and repository checked on 25 October 2026 · added on 1 July 2024

How to install LivePortrait

1. Install git, conda and FFmpeg on your system. 2. Clone the repository: git clone https://github.com/KwaiVGI/LivePortrait then enter the folder. 3. Create and activate a conda environment: conda create -n LivePortrait python=3.10 then conda activate LivePortrait. 4. Install dependencies: pip install -r requirements.txt (or requirements_macOS.txt on Apple Silicon). 5. Download the pretrained weights from HuggingFace: huggingface-cli download KlingTeam/LivePortrait --local-dir pretrained_weights. 6. Never animate another person's face without their explicit consent.

git clone https://github.com/KwaiVGI/LivePortrait && cd LivePortrait && conda create -n LivePortrait python=3.10 && conda activate LivePortrait && pip install -r requirements.txt

In the same category

Whole category Video →