Orange
A visual data mining and machine learning toolbox where analyses are built as workflows of widgets, so users need little programming. Built in Python.
- GitHub stars
- 5.7k
- Last commit
- 4 days ago
- Latest release
- 3.40.0
- Self-hosted
- Yes

Orange is a data mining and visualization toolbox aimed at both beginners and experienced analysts. Instead of writing code, users assemble analyses as workflows on a canvas, which hides the underlying mechanics and exposes intuitive concepts. The project's stated goal is that anyone who owns data should be able to explore it.
The toolbox covers interactive data visualization along with classic machine learning methods such as classification, regression, clustering, decision trees and random forests. It is written in Python on top of scikit-learn, NumPy, SciPy and pandas, and extra functionality is added through add-ons installed from the menu bar; developers can write their own widgets from an example template.
Windows and macOS users can download a standalone installer, while Linux users and developers can install with conda, pip or uv, and winget is available on Windows. The project is open source, welcomes new widgets and contributions, and has a community on Discord.
Key features
- Visual workflow canvas built from widgets
- Interactive data visualization
- Classification, regression and clustering
- Decision trees and random forests
- Add-ons installable from the menu bar
- Custom widgets written in Python
- Standalone installers for Windows and macOS
Pricing: Free and open source.
