Skip to content

Installation Options

Khiops supports a diversified set of installation options, to meet different needs:

  • Khiops Python Library:
  • Khiops Application: for advanced data analytics with just a few clicks using a graphical user interface. This application is also the basis for easy integration into different systems (all programming languages, docker, servers, etc.).
  • Khiops Visualization Application: for intuitive visualization of all analysis results
  • Khiops Native Interface (KNI): to deploy Khiops models with a lightweight shared library.

We support Python from 3.8 to 3.12 and the following operating systems:

  • Windows 10 or later
  • Ubuntu 20 and 22 (LTS)
  • Debian 10, 11 and 12
  • Rocky Linux 8 and 9
  • macOS 12 or later, only via Conda.

The Kaggle Notebooks and Google Colaboratory environments are supported. To benefit from Khiops on these environments, users are encouraged to install the Khiops Conda package, which has been tested in these environments.

For other platforms, please Contact Us.

Install the Khiops Python Library Using Conda Recommended

The Conda package contains all the necessary components.

conda install -c conda-forge -c khiops khiops
conda install -c khiops khiops
conda install -c conda-forge -c khiops khiops

We support Python from 3.8 to 3.12. If you are using a different version of Python in your current environment, we recommend visiting our Conda Installation Page for instructions on creating a new environment tailored for Khiops:

See the Conda Installation Page

Run Khiops with Jupyter Docker Stacks

For a quick and easy way to get started with Khiops, you can use our Docker container.

docker pull khiopsml/khiops-notebook

See the Docker notebooks Installation Page

Install the Khiops Application

This version contains a standalone Graphical User Interface (GUI).

The Khiops installer automatically installs the Khiops application, all its dependencies, plus the Khiops samples and the Khiops Visualization application:

CODENAME=$(lsb_release -cs) && \
TEMP_DEB_CORE="$(mktemp)" && \
TEMP_DEB_KHIOPS="$(mktemp)" && \
wget -O "$TEMP_DEB_CORE" "${CODENAME}.amd64.deb" && \
wget -O "$TEMP_DEB_KHIOPS" "${CODENAME}.amd64.deb" && \
sudo dpkg -i "$TEMP_DEB_CORE" "$TEMP_DEB_KHIOPS" || sudo apt-get -f -y install && \

Currently, our packages are released on GitHub. In the coming weeks, we will transition to official repositories.

sudo yum update -y && sudo yum install wget -y && \
CENTOS_VERSION=$(rpm -E %{rhel}) && \
TEMP_RPM="$(mktemp).rpm" && \
TEMP_RPM_KHIOPS="$(mktemp).rpm" && \
wget -O "$TEMP_RPM" "${CENTOS_VERSION}.x86_64.rpm" && \
wget -O "$TEMP_RPM_KHIOPS" "${CENTOS_VERSION}.x86_64.rpm" && \
sudo yum install "$TEMP_RPM" "$TEMP_RPM_KHIOPS" -y && \

See the Khiops Desktop Installation Page

Install the Khiops Visualization Applications

See the Khiops Visualization Page Try our Interactive Demo

Install the Khiops Python library using Pip 🛠 For Advanced users 🛠

Using pip for installation is best suited for experienced users, especially those integrating Khiops into specific environments or requiring advanced configuration. This method offers flexibility, but requires familiarity with operating systems specifics, Python-based environments and dependency management.

See the Pip Installation Page