OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Visomaster Fusion

Visomaster Fusion is a GPL-3.0-licensed desktop application for AI face swapping, enhancement and editing of images, videos and live webcam feeds. Its graphical workflow includes model controls, multi-face editing, restoration tools, batch jobs, VR180 support and GPU-accelerated inference.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Visomaster Fusion — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a local desktop workflow for face swaps or enhancement across images, videos and live feeds, or need controls for multi-face editing, video segments and batch jobs.

Who it is for

  • Creators editing faces in still images or video.
  • Users preparing webcam or virtual-camera face-processing workflows.
  • Video editors who need per-frame controls, segment renders or unattended batch jobs.
  • Developers looking to inspect or adapt a Python-based, GPL-3.0 face-editing application.

Documented uses

  • Swap faces in images, video clips or webcam feeds, including workflows with multiple faces.
  • Adjust likeness, keypoints, expressions and pose, then apply masks or face restoration.
  • Save per-frame settings, review render-sensitive frames and produce image, video or multi-segment outputs.
  • Save workspaces and run batches unattended through the Job Manager.
Before you choose

Requirements & limits

The documented portable setup targets 64-bit Windows 10 or 11. It needs internet access on first launch to download dependencies and models. The project identifies its license as GPL-3.0.

  • The app is documented for Windows 10 and 11, 64-bit.
  • An Nvidia GPU is recommended; CPU use is available but considerably slower.
  • The README lists 6 GB VRAM for basic use and recommends 8–12 GB or more for heavier work.
  • The portable launcher downloads runtime components and models on first launch; allow 20–30 GB of disk space.
  • The application offers CUDA, TensorRT, TensorRT-Engine and CPU providers, with TensorRT listed as the default.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and start VisoMaster Fusion in a managed Python environment

Set up the non-portable Windows installation, including its Python environment, requirements, models, and FFmpeg, then start the application.

Before you start

  • A Windows system; the README specifies Windows 10 or Windows 11, 64-bit.
  • Python 3.12 environment management through Anaconda or uv.
  • Internet access for retrieving the repository, packages, models, and FFmpeg.

Steps

  1. Clone the repository with the command below and change into the VisoMaster-Fusion directory.

    Documented command
    git clone https://github.com/VisoMasterFusion/VisoMaster-Fusion

  2. Create and activate a Python 3.12 environment using either the documented Anaconda commands or the uv commands; install uv with pip when using Anaconda.

  3. Install requirements with the command below, then download required models with the command below.

    Documented command
    uv pip install -r requirements_cu13.txt
    Documented command
    python download_models.py

  4. Install FFmpeg using one of the README's Windows options: the listed winget command, or the specified FFmpeg archive with its bin directory added to Windows PATH.

  5. Start the app by opening Start.bat on Windows, or run the command below in the activated environment from the repository directory.

    Documented command
    python main.py

Expected outcome

The documented non-portable setup installs dependencies and models and provides a way to start the application.

Guide limitations

  • This path is intended for users who want to manage the Python environment themselves; the README recommends the portable version for most users.
  • CPU processing is supported but much slower than GPU processing.
  • These steps cover setup and startup, not a detailed editing workflow.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
VisoMasterFusion/VisoMaster-Fusion · Official homepage title · retained in its source language
How we review sourcesSuggest a correction