Get Started With Intel® Distribution for Python*

Use these resources to get up to speed fast.

You can have several versions of Python on your system. Conda* also lets you manage multiple Python environments. By installing Intel Distribution for Python in a conda environment, you ensure that your system installation of Python will not be affected.

Virtualenv does not copy a required library, libpython, into the virtual directory. For details and workarounds, see this discussion on GitHub*.


We recommend managing environments with conda instead.

To set environment variables with Bash*, do one of the following:

For macOS* and Linux*:


export CC=icc


export LD_SHARED=”icc -shared”


For Windows*:


set CC=icl


set LD=xilink

To set environment variables with Bash, do one of the following:


For macOS and Linux:

export CC=icc

export LD_SHARED=”icc -shared”


For Windows:

set CC=icl

set LD=xilink

export CXX=icpc

Not directly. Many functions in NumPy and SciPy are implemented with Intel MKL functions. Since Intel Distribution for Python has the same shared libraries and functions as Intel MKL, you can build your own C extensions that link to the functions. For details, see the Intel MKL Documentation.

Yes. See INTEL_PYTHON_EULA and redist.txt in the install directory for details. This document is downloaded as part of your installation.

The –devel and –static packages can be used by developers building applications that require linking to Intel runtimes included in Intel® Distribution for Python*. For example, a developer may choose to build their own NumPy package with Intel MKL routines.

If you plan to compile against library versions in your Intel Parallel Studio XE installation, you should use the –devel packages included therein.

Starting with Intel Parallel Studio XE 2019, a local conda channel is included at the Intel Parallel Studio XE root (<psxe_installdir>/conda_channel) that contains conda packages of all Intel® Performance Libraries. Installing the libraries (with conda install mkl-devel) from that location links the libraries in Intel Parallel Studio XE into the Python environment directly from the canonical installation. This method ensures consistent library versions. We’ve added this location to the .condarc file in the intelpython env (<psxe_installdir>/intelpython), so it should work out of the box.

Intel MKL, Intel® Integrated Performance Primitives (Intel® IPP), and Intel® Data Analytics Acceleration Library (Intel® DAAL) development versions have been included in various forms for developer use. They are located in the <python_root>/pkgs directory and come as conda packages with the following naming scheme (where PROD is the product name):

PROD: shared libraries needed at runtime

PROD-include: header files

PROD-devel: shared libraries plus header files, used for building using dynamic linking

PROD-static: shared libraries plus header files, used for building using static linking

Intel® Tech.Decoded


Gain insight on what's ahead with software—from parallel programming and HPC to data science and computer vision. Access online training, webinars, and quick-tips to help you learn and master Intel® Software Development Products.

Product and Performance Information


Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804