There are two options how to install cubes: basic common installation - recommended mostly for users starting with Cubes. Then there is customized installation with requirements explained.

Basic Installation

The cubes has optional requirements:

  • SQLAlchemy for SQL database aggregation browsing backend (version >= 0.7.4)
  • Flask for Slicer OLAP HTTP server


If you never used Python before, you might have to get the pip installer first, if you do not have it already.


The command-line tool Slicer does not require knowledge of Python. You do not need to know the language if you just want to serve OLAP data.

You may install Cubes with the minimal dependencies,

pip install cubes

with certain extras (html, sql, mongo, or slicer),

pip install cubes[slicer]

or with all of the extras.

pip install cubes[all]

If you are developing cubes, you should install cubes[all].

Quick Start or Hello World!

Download the sources from the Cubes Github repository. Go to the examples/hello_world folder:

git clone git://
cd cubes
cd examples/hello_world

Prepare data and run the OLAP server:

slicer serve slicer.ini

And try to do some queries:

curl "http://localhost:5000/cube/irbd_balance/aggregate"
curl "http://localhost:5000/cube/irbd_balance/aggregate?drilldown=year"
curl "http://localhost:5000/cube/irbd_balance/aggregate?drilldown=item"
curl "http://localhost:5000/cube/irbd_balance/aggregate?drilldown=item&cut=item:e"

Customized Installation

The project sources are stored in the Github repository.

Download from Github:

git clone git://


cd cubes
pip install -r requirements.txt
pip install -r requirements-optional.txt
python install


The requirements for SQLAlchemy and Flask are optional and you do not need them if you are going to use another kind of backend or don’t going to use the Slicer server.