Choosing a particular backend can be done using`af.set_backend(name)` where name is either "_cuda_", "_opencl_", or "_cpu_". The default device is chosen in the same order of preference.

ArrayFire can be installed from a variety of sources.[Pre-built wheels](https://repo.arrayfire.com/python/wheels/3.8.0/) are available for a number of systems and toolkits.Wheels for some systems areavailable on PyPI.Unsupported systems will require separate installation of the ArrayFire C/C++ libraries and only the python wrapper will be installed in that case.

ArrayFire can be installed from a variety of sources.[Pre-built wheels](https://repo.arrayfire.com/python/wheels/3.8.0/) are available for a number of systems and toolkits.These will include a distribution of the ArrayFire libraries. Currently, only the python wrapper isavailable on PyPI.Wrapper-only installations will requireaseparate installation of the ArrayFire C/C++ libraries.

You can get the ArrayFire C/C++ library from the following sources:

-[Download and install binaries](https://arrayfire.com/download)

-[Build and install from source](https://github.com/arrayfire/arrayfire)

**Install the last stable version:**

**Install the last stable version of python wrapper:**