Python and uv

Use the Python installation path when you want Toposync to run directly on the host system instead of Docker.

Toposync documentation uses uv because it makes Python version management, virtual environments, and package installation more predictable across Linux, macOS, and Windows.

Choose a Python path

System	Recommended guide
Linux or macOS	Python on Linux and macOS
Windows desktop or server	Python on Windows
Remote processing server on Linux or macOS	Processing server on Linux and macOS
Persistent processing server on Windows	Processing server as a Windows service

Default recommendation

Use Python 3.12 through uv, create a virtual environment, then install the Toposync package you need:

uv python install 3.12
uv venv .venv --python 3.12

After that, follow the guide for your operating system.

Package choices

Package	Use when
toposync	You want the default CPU product
toposync-streaming	You need the streaming extension
toposync-vision-directml	You want Windows GPU acceleration through DirectML
toposync-vision-cuda	You want NVIDIA CUDA acceleration

Start with toposync. Add streaming or GPU packages later when there is a concrete need.

For compatibility details, see Compatibility.