napari
Fast, interactive, multi-dimensional image viewer for Python.
Overview
napari displays n-dimensional images (2D/3D/4D+) with layered overlays for labels, points, shapes, and tracks. It's often used as the front-end for Python-based analysis workflows and as a host for plugins that wrap other tools (Cellpose, micro-sam, StarDist, and many more).
Key features
- N-dimensional viewing with synchronized slicing
- Layer system for images, labels, points, shapes, surfaces, tracks, vectors
- Plugin ecosystem via napari-hub
- Scriptable from Jupyter, IPython, or standalone scripts via the Python API
Installation
See Python environments for package manager setup (pixi or conda).
Recommended if you want napari alongside a specific tool. The AI_tools_pixi repository bundles napari into the environments where it's actually used:
AI_tools_pixi/omero— napari + napari-omero for browsing and annotating OMERO images.AI_tools_pixi/micro_sam— napari + micro-sam + trackastra + napari-omero.AI_tools_pixi/stardist— napari + StarDist.
Clone the repo, cd into the folder, then pixi install and pixi run napari.
Standalone napari (pixi).
mkdir napari && cd napari
pixi init
pixi add python=3.11 napari pyqt
pixi run napari
Standalone napari (conda).
conda create -n napari -c conda-forge python=3.11 napari pyqt
conda activate napari
napari
For pip or other install options, see the official install guide.
Official documentation
Learning resources
Written
- Getting started tutorial
- napari-hub — browse plugins by task (segmentation, tracking, file IO, etc.).
- napari tag on Image.sc — community Q&A.