LibreAI / Video / LivePortrait
// video · KlingAIResearch/LivePortrait

LivePortrait

LivePortrait is an open-source video project by KlingAIResearch (KlingAIResearch/LivePortrait on GitHub). It has 19,160 GitHub stars and 1,992 forks, is released under the Custom (check repo) license, and was last updated on GitHub on 1 June 2026.

Bring portraits to life!

★ 19.2k stars⑂ 2.0k forksCustom (check repo)updated 2026-06
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
// project facts

Project facts

GitHub stars
19,160
Forks
1,992
License
Custom (check repo)
Last updated on GitHub
1 June 2026
Category
Video
Listing
Full install guide
Official site
Official site
// how it works

How it works

  1. Pick your task

    Generation, restoration, upscaling, face swap, or lip sync — each project here specializes, so start from the job to be done.

  2. Get the weights

    Download the model files the project needs; most READMEs link exact versions and where to place them.

  3. Set up the environment

    Python environments or Docker images are the norm — follow the install section for your GPU.

  4. Run and iterate

    Feed in prompts or source media, tune the settings, and re-run until the result lands.

// features

Why people love it

Open weights

Real model files you download and run — no API key, no per-second billing.

Local inference

Everything happens on your machine; footage never uploads anywhere.

Specialized pipelines

Each tool focuses on one job and does it well, instead of a shallow all-in-one.

Active research roots

Many of these ship alongside papers and keep absorbing new techniques.

Community workflows

Shared settings, presets, and integrations (especially around ComfyUI) multiply what one tool can do.

Commercial-friendly options

Several use permissive licenses — always confirm the model license for your use case.

Advertisement
// run it

Run it yourself

  1. Step 1 — Clone the repo

    Download the source and move into the project folder.

    git clone https://github.com/KlingAIResearch/LivePortrait.git
    cd LivePortrait
  2. Step 2 — Create an environment

    Isolate dependencies so nothing clashes with your other projects.

    python -m venv .venv
    source .venv/bin/activate  # Windows: .venv\Scripts\activate
  3. Step 3 — Install and launch

    Install the requirements, then follow the README's launch command (it varies per project).

    pip install -r requirements.txt
// use cases

What you can build

  • Restore and upscale old or low-res footage.
  • Animate portraits with lip-synced speech.
  • Generate clips from text with open video models.
  • Swap or enhance faces in existing video.
Advertisement
// faq

Questions, answered

What GPU do I need?

It varies widely — upscalers run on modest cards, while full video diffusion wants serious VRAM. Check each project's docs.

Are the outputs commercially usable?

Often yes for the code, but model weights carry their own licenses — read both.

Who maintains LivePortrait?

It is an open-source project at KlingAIResearch/LivePortrait with 1,992 forks. Check the repo's contributors and recent commits to see how active maintenance is.

How is quality vs commercial tools?

The gap keeps shrinking; for restoration and upscaling, open tools already match or beat paid ones.

Do these run on Mac?

Some do via MPS or CPU, but NVIDIA on Linux/Windows is the best-supported path.

Where do I get help?

GitHub issues and Discussions on each repo, plus Discord communities for the popular ones.

// keep exploring

Similar tools

Video

FaceFusion

Industry leading face manipulation platform

★ 30.1k starsCustom (check repo)
View guide →
Video

Video2X

A machine learning-based video super resolution and frame interpolation framework. Est. Hack the Valley II, 2018

★ 21.9k starsAGPL-3.0
View guide →
Video

Wan2.1

Wan: Open and Advanced Large-Scale Video Generative Models

★ 17.1k starsApache-2.0
View guide →
Video

editor-pro-max

Describe videos in plain language — Remotion plus Claude Code renders the MP4.

★ 258 starsCustom
View guide →
Video

youtube_summarizer

Self-hosted video: a modern next.js-based tool for ai-powered youtube vide.

★ 217 starsMIT
View guide →
Video

GhostCut_Remake_Video

Remix videos: strip subtitles, rewrite captions, AI re-dub.

★ 206 starsCustom
View guide →
// alternatives

Alternatives in Video

The most-starred other Video projects in the LibreAI catalogue — the realistic shortlist if LivePortrait is not the right fit.

  • yt-dlp★ 195.3k

    A feature-rich command-line audio/video downloader

  • Deep-Live-Cam★ 96.9k

    real time face swap and one-click video deepfake with only a single image

  • obs-studio★ 76.9k

    OBS Studio - Free and open source software for live streaming and screen recording

  • FFmpeg★ 64.7k

    Mirror of https://git.ffmpeg.org/ffmpeg.git

  • OpenMontage★ 62.7k

    World's first open-source, agentic video production system. 12 production pipelines, 100+ tools, 700+ agent skill and production-knowledge files. Turn your A…

// more in video

More in Video

Nearby Video projects in the catalogue, closest in stars to LivePortrait — keep exploring the shelf either side of it.

  • pyvideotrans★ 19.3k

    Translate videos end to end - speech recognition, translated subtitles and dubbing in one app

  • hypit★ 19.0k

    Clone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-roll, ship 100 variants in one command, and get …

  • gallery-dl★ 19.9k

    Command-line program to download image galleries and collections from several image hosting sites

  • VideoLingo★ 18.7k

    Netflix-level video subtitles and dubbing - a one-click AI subtitle team

  • blender★ 20.7k

    Official mirror of Blender

  • FramePack★ 17.3k

    Lets make video diffusion practical!

Get LivePortrait

Free and open — grab the source, star the repo, and start creating.