Startup Stash is one of the world's largest online directory of tools and resources for startups
Free, open source visual tool for machine learning and data visualization.
Orange is a free, open source desktop app for machine learning and data visualization, made by the Bioinformatics Lab at the University of Ljubljana. You build an analysis by connecting widgets on a canvas, with no code needed.
Orange was first released in 1996 and is written in Python, Cython, C++ and C. The current version is 3.40.0, released in December 2025. Add-ons cover text mining, network analysis, time series, spectra and gene expression data. People often search for an online version, but the official site only offers downloads for Windows, macOS and installs through pip or conda. There is no browser version.
Bottom line: Orange is a strong pick for teaching, learning and quick exploratory analysis at no cost. If you need to work in a browser or share live analyses with a team, look at hosted tools.
Drag widgets onto a canvas and connect them to load data, train models and compare results, with no coding.
Box plots, scatter plots, decision trees, hierarchical clustering, heatmaps, MDS, t-SNE and linear projections.
Extra widgets for text mining, network analysis, machine learning fairness, time series, spectra and gene expression.
Use the Orange data mining library from Python code when the visual canvas is not enough.
| Plan | Price | What you get |
|---|---|---|
| Orange desktop app | Free | Full app for Windows (installer or portable) and macOS (Apple silicon or Intel). |
| pip or conda install | Free | Install with pip install orange3 or from conda-forge, for Linux and other systems. |
| Donations | Optional | The project asks for donations to fund fixes, new features and teaching material. |
Orange is open source under the GNU GPL, version 3 or later. There are no paid plans.