ai-agent

Faceswap: AI Face Replacement

Name: Faceswap: AI Face Replacement
Author: deepfakes

Get deepfake software for AI face swap, ideal for founders in AI and media. 55k+ GitHub stars

57,446 stars13,499 forksPythonHealth Score 9/10Updated 8/5/2026100% free · open source

What it does

Faceswap is a deepfake software that uses AI to swap faces in images and videos, allowing founders in AI and media to create realistic face replacements.

Install / run

git clone https://github.com/deepfakes/faceswap.git && cd faceswap && pip install -r requirements.txt

When to use it

•Creating realistic face replacements for film or video production
•Developing AI-powered face manipulation tools for social media or entertainment
•Generating synthetic faces for data augmentation in machine learning models

Quick start

1Create a new project using the command `python scripts/train.py -m new -p /path/to/project`
2Add faces to the project by running `python scripts/extract.py -i /path/to/images -o /path/to/output`
3Configure the model settings in the `config.json` file
4Train the model using the command `python scripts/train.py -m train -p /path/to/project`
5Use the trained model to swap faces with the command `python scripts/convert.py -i /path/to/input -o /path/to/output -m /path/to/model`

Ready-to-paste prompt

python scripts/convert.py -i /path/to/input.jpg -o /path/to/output.jpg -m /path/to/model -f /path/to/face.jpg

Heads up: Requires a compatible version of Python (3.8 or later) and a significant amount of GPU memory (at least 8 GB) to run efficiently

Saves to your device

Topics

deep-face-swap

deep-learning

deep-neural-networks

deepface

deepfakes

deeplearning

face-swap

faceswap

fakeapp

machine-learning

myfakeapp

neural-nets

neural-networks

openfaceswap

How Faceswap: AI Face Replacement works

Codeflow

Free to inspect

Faceswap: AI Face Replacement is a very large Python project (~93k lines across 283 code files, plus 40 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, GPL-3.0 license, has a test suite.

Size

Very large codebase

~93k lines · 283 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Python

No API keys detected

Python 99%Shell 1%

Where to start reading