Installation

Requirements

  • Blender 5.1.0+
  • Bundled Python wheels target Python 3.13 (the interpreter shipped with Blender 5.1.x).
  • Supported platforms: Linux x64, Windows x64, macOS ARM64 (Apple Silicon).
  • Internet access is required for geospatial features (geocoding, OSM networks, maps, DEM) and SQL database connections.

Every dependency ships as a wheel listed in blender_manifest.toml, including the project’s own scigraphs-core and scigraphs-engine, so no manual pip install is needed.

Install from a release

  1. Download the platform zip for your system from the GitHub releases / Actions artifacts.
  2. Open Blender, then go to Edit ▸ Preferences ▸ Add-ons.
  3. Select Install from Disk, then choose the downloaded zip.
  4. Enable the SciGraphs add-on if it is not already active.

Install the development version

Open GitHub Actions, pick the latest successful run, and download the artifact for your platform. Extract it to obtain the actual extension zip, then install that zip as above.

Build from source

git clone https://github.com/SciBlend/SciGraphs.git
cd SciGraphs
./scripts/fetch_wheels.sh   # fetch third-party wheels, build the project's own
./build_extension.sh        # build per-platform zips and install for your OS

The wheels are not tracked in git, so fetch_wheels.sh has to run first. It downloads the Python 3.13 third-party wheels for every platform from the constraints/ files, and builds scigraphs-core and scigraphs-engine from core/ and engine/ in the checkout. build_extension.sh then runs Blender’s extension build --split-platforms to produce per-platform zips in dist/ and installs the one matching your operating system.

Opening the panels

After enabling the add-on, open the 3D Viewport sidebar with N and look for the SciGraphs tabs:

  • SciGraphs for the core combinatorial workflow (Data, Layout, Algorithms, Analysis, Visualization, Export, Reproducibility).
  • OSMnx for street-network download and analysis.
  • City2Graph for urban morphology, transport, and heterogeneous graphs.
Tip

If the tabs do not appear, confirm the add-on is enabled and that you are in the 3D Viewport (View3D) editor. The sidebar tab labels are docked vertically along the right edge of the viewport.

Native bindings

Two native bindings are built and maintained by the project author and bundled as platform-specific wheels: scigraphs-utils powers the Graphviz layouts (dot, neato, fdp, sfdp, twopi, circo, osage, patchwork, including native 3D variants), and pysurprise provides the SurpriseMe community-detection algorithms and the Surprise quality metric.

Back to top