VisoMaster System Requirements
VisoMaster runs 100% locally on your Windows PC and does all of its AI processing on your NVIDIA GPU. Here is exactly what you need for smooth face swapping.
| Minimum | Recommended | |
|---|---|---|
| Operating system | Windows 10 (64-bit) | Windows 11 (64-bit) |
| GPU | Dedicated NVIDIA GPU, 6GB VRAM | NVIDIA RTX 40 / 50 Series, 8GB+ VRAM |
| System RAM | 8GB | 16GB or more |
| Disk space | ~20GB free | SSD with 20GB+ free |
| Internet | Only for downloading the package | — (all processing is 100% local) |
GPU Support
- NVIDIA RTX 50 Series (5070, 5070 Ti, 5080, 5090, 5090 D — desktop and laptop): fully supported with native Blackwell (sm_120) builds. Hitting a CUDA error on launch? See the RTX 50 fix guide.
- NVIDIA RTX 40 / 30 / 20 Series and GTX 16 Series: supported, provided the card has at least 6GB of VRAM.
- AMD and Intel GPUs: not currently supported — VisoMaster relies on CUDA and TensorRT acceleration.
More VRAM directly improves the experience: 6GB cards should run one restorer model with threads set to 1, while 8GB+ cards can enable TensorRT engines for significantly faster previews and exports. See the performance tuning section of the VisoMaster tutorial for per-GPU settings.
Why No Python or CUDA Install?
The all-in-one package bundles a pre-configured runtime with all dependencies included. As long as your hardware meets the requirements above and you have a current NVIDIA driver, VisoMaster runs straight from the extracted folder — no environment setup of any kind.
Your PC Meets the Requirements?
Get the all-in-one package and make your first face swap in minutes.
Download VisoMaster