Databench is a Python data analysis tool using Tornado and WebSockets that you can install with:

$ pip install databench

Live demos are at and a user guide and API documentation at Databench is made available under the MIT license.


Databench provides the executables scaffold-databench and databench, Python modules for the backend and a JavaScript library for the frontend. scaffold-databench helloworld creates an analysis template called helloworld in the current working directory. Running databench creates a local web server which you can access at http://localhost:5000. A good way to start is to jump right into Quickstart.

Some features are shown in the live demos. They do not include examples with matplotlib, parallelization or database interfaces (like asynchronously subscribing to a Redis channel) but those examples are available in the databench_examples repository.

Preview of flowers demo. Preview of bag-of-chars demo. Preview of simplepi demo.