Please install/save everything locally to prevent any conflicts with other libraries. Here we introduce two variables

  • RAISIM_DIR: the directory where you cloned raisimLib

  • LOCAL_INSTALL: install directory where you install exported cmake libraries (e.g., ~/raisim_build)

You can add them to your environment variables or simply replace them by the path you want in the following instructions on this page.


You have to install the following packages before using raisim

  • eigen library

  • cmake > 3.10

  • For windows users, visual studio 2019 (we only support 2019 but it will probably work with 2015 or above). Make sure that you install the C++ module as well by checking the corresponding checkbox during install.

RaiSim includes many open-source libraries. See the COPYING file for the full list. Source code of Eigen and Pybind11 are included in the thirdParty directory.

RaiSim Install

Clone raisim from

The cloned repo is already a set of installed raisim packages. Under raisimLib/raisim/<OS-TYPE>, you will find the installed cmake packages.

To use raisim in your project, you can simply add the cmake package path and the shared library path to their corresponding environment variables as following

Add the following lines to your ~/.bashrc file


This allow the linux linker to find the raisim shared libraries. To let your project CMake know where to find raisim, simply pass an argument -DCMAKE_PREFIX_PATH=$WORKSPACE/raisim/linux.

Building RaiSim Examples and Installing RaisimPy

You can use the CMakeLists.txt in the raisimLib directory to build raisim examples and other modules. The following options are available

  • RAISIM_EXAMPLE : Compile C++ RaiSim examples

  • RAISIM_MATLAB : Compile raisimMatlab (compiled binary is also provided). You need MATLAB for this option

  • RAISIM_PY : Compile raisimPy. When installed, it will copy raisimpy files to your python site-packages directory

You can generate build files using CMake as following

To build raisimPy for the correct python version, activate your conda environment before calling cmake.

mkdir build
cd build
make install -j4


For Linux users To use (vulkan version) RaiSimUnity in Linux, you need to install minizip, ffmpeg and vulkan. If you are going to use raisimGym, install the recommended version of gpu driver by pytorch. If not, install the latest version.

To install minizip and ffmpeg,

sudo apt install minizip ffmpeg

Vulkan installation depends on your OS distribution. You can easily find instructions online.

Ubuntu 22.04 users should run the following command before running RaisimUnity.

sudo ln -s /usr/lib/x86_64-linux-gnu/ /usr/lib/x86_64-linux-gnu/

For other versions of Ubuntu, the symlink is provided by Ubuntu automatically.

If you still cannot use raisimUnity, this probably means that your driver does not support vulkan. In that case, you should use raisimUnityOpengl. It only supports minimalistic graphics.

Activation Key

Rename the activation key that you received by email to activation.raisim. Save that file in <YOUR-HOME-DIR>/.raisim. In Linux and Mac, this is /home/<YOUR-USERNAME>/.raisim. In Windows, this is C:\Users\<YOUR-USERNAME>\.raisim (You might not be using C as your home directory).

RaiSim will also check the path you set by raisim::World::setActivationKey(). If the file is not found, it will search in the user directory, where you saved your activation.raisim file.


The built examples are stored in examples directory. In Windows, use powershell to run the examples, instead of manually clicking the icons. If you made a mistake during installation, it will give you an error message.

Make sure that you run raisimUnity executable in raisimUnity/<OS>/RaiSimUnity before you run the examples.