Get started
Installation
VisoLabel ships as a native desktop app for Windows, macOS, and Linux. Download the installer for your platform, run it, and you're ready to label. No Docker, no YAML, no driver wrangling.
System requirements
| Component | Minimum | Recommended |
|---|---|---|
| Operating system | Windows 10, macOS 12, Ubuntu 20.04 | Windows 11, macOS 14, Ubuntu 22.04 |
| RAM | 8 GB | 16 GB or more |
| Disk | 5 GB free | 20 GB+ (for datasets and models) |
| GPU (for local training) | Optional — CPU works for annotation | NVIDIA GPU with 8 GB+ VRAM |
No GPU? You can still annotate with SAM2 and SAM3 and train in the cloud via the Google Colab pipeline.
Download
Grab the latest installer from the download page and pick your platform below.
Windows
- Download the
VisoLabel-Setup.exeinstaller. - Double-click it and follow the prompts.
- If Windows SmartScreen appears, choose More info → Run anyway (the app is code-signed but new releases take time to build reputation).
- Launch VisoLabel from the Start menu.
macOS
- Download the
VisoLabel.dmgfile. - Open it and drag VisoLabel into your Applications folder.
- On first launch, right-click the app and choose Open to approve it in Gatekeeper.
Linux
VisoLabel is distributed as an AppImage and a .deb package.
# AppImage — works on most distributions
chmod +x VisoLabel.AppImage
./VisoLabel.AppImage
# Debian / Ubuntu
sudo dpkg -i visolabel_amd64.debFirst launch
The first time you open VisoLabel it downloads the SAM2 and SAM3 model weights so assisted annotation works fully offline afterwards. This is a one-time download of a few hundred megabytes.
- Open VisoLabel.
- Wait for the model weights to finish downloading (progress is shown).
- Optionally sign in to your account to sync your license and unlock Pro features.
Verify the install
To confirm everything works, create a test project and import a single image — the steps are covered in the Quickstart. If SAM2 assisted clicks produce a polygon, your install is good to go.