The scripting interface of OVITO lets you

  • automate data visualization and analysis steps,

  • integrate OVITO’s data I/O, analysis and rendering capabilities into custom workflows or Python programs,

  • extend OVITO capabilities by developing new modifiers or viewports layers that integrate into the graphical user interface.

The script programming interface described in this manual consists of a collection of Python functions and classes, which expose OVITO’s internal data model, data pipeline framework and rendering capabilities.

Automation scripts

A common use case for OVITO’s programming interface is automating laborious tasks. Python scripts don’t require any user interaction and can (repeatedly) perform sequences of actions that you would otherwise have to carry out by hand in the graphical application. The following example script loads an atomic structure from a simulation file, selects all hydrogen atoms, deletes them, and writes the resulting dataset back to an output file:

from import import_file, export_file
from ovito.modifiers import SelectTypeModifier, DeleteSelectedModifier

pipeline = import_file('')
pipeline.modifiers.append(SelectTypeModifier(property='Particle Type', types={'H'}))
export_file(pipeline, '', 'xyz', columns=['Particle Type', 'Position.X', 'Position.Y', 'Position.Z']])

Scripts like this example can be easily created with OVITO Pro’s built-in code generator in the graphical user interface. You can then execute the Python script from the system terminal using ovitos, the preconfigured Python interpreter shipping with the OVITO Pro desktop application:


Alternatively, you can execute the script – like any other Python program – using your Python interpreter of choice:


For this to work, you first need to install the ovito module in your Python interpreter as will be explained later.

Running scripts within the desktop application

A Python script can also be executed right within the context of the running OVITO Pro desktop application. This option can be useful especially during script development, because it allows you to directly observe what your script does and how the final outcome, for example the data pipeline created by the script, looks like.

Use the Run Script File function found in the File menu of OVITO Pro to execute an existing Python script file. Note that, in this case, all script actions will be executed within the context of the current program session – as if you would perform them by hand. Alternatively, you can use the --script command line option when invoking OVITO from the system terminal to execute a given Python script right after application startup.

User-defined modifiers and viewport overlays

OVITO’s scripting interface furthermore enables you to develop new types of modifiers that can manipulate or analyze data in ways not covered by any of the built-in modifiers of the program. So-called Python script modifiers (see User-defined modifiers section for more information) participate in the data pipeline system of OVITO and behave just like the built-in modifiers from a user’s perspective. Such a user-defined modifier essentially is a Python function written by you that gets executed automatically by OVITO whenever the data pipeline is evaluated.

Furthermore, you can write custom viewport layers. A script viewport layer is a user-defined Python function that gets called by OVITO every time a viewport image is being rendered. This allows you to enrich rendered images or movies with custom text and graphics, and include additional information such as data plots that are dynamically generated from the simulation data.

Installing the ovito module in your Python interpreter

To make OVITO’s scripting capabilities available to scripts running in your system’s Python interpreter, you first need to install the ovito module. It may be downloaded from the PyPI repository and installed using the pip install ovito command. For further instruction please refer to

If you are using the Anaconda or Miniconda environments, you should install the ovito Python module from our channel as follows:

conda install --strict-channel-priority -c -c conda-forge ovito=3

The conda-forge channel provides additional dependencies required by OVITO. The ovito conda package includes both the graphical desktop application OVITO Pro and the ovito Python module.

OVITO’s Python interpreter

OVITO program packages include a preconfigured Python interpreter named ovitos that can execute scripts written in the Python 3.x language. ovitos already has the ovito module and all other dependencies preinstalled that are required to run scripts. You can execute a Python script from the terminal of your operating system:

ovitos [-o file] [-g] [] [args...]

The ovitos executable is located in the bin/ subdirectory of OVITO for Linux, in the directory of OVITO for macOS, and in the main application directory on Windows systems (look for ovitos.exe). It should not be confused with ovito, which is the graphical desktop application.

Let’s use a text editor to write a simple Python script file named

import ovito
print("Hello, this is OVITO %i.%i.%i" % ovito.version)

You can execute the script file from a Linux terminal as follows:

me@linux:~/ovito-3.0.0-x86_64/bin$ ./ovitos
Hello, this is OVITO 3.0.0

The ovitos script interpreter is a console program without a graphical user interface. This enables you to run scripts on remote machines or computing clusters that don’t possess a graphical display. ovitos behaves like a regular Python interpreter. Any command line arguments following the script name are passed to the script via the sys.argv variable. Furthermore, it is also possible to start an interactive interpreter session by invoking ovitos without any arguments.

Preloading session state

The -o command line option tells ovitos to load an .ovito session state file before executing the script. This allows you to preload an existing data pipeline or visualization setup that you have previously prepared using the graphical version of OVITO. All script actions will subsequently be performed in the context of this preloaded session. This can save you programming work, because things like modifiers and the camera setup already get loaded from the state file and you don’t need to set everything up programmatically in the script anymore.

Graphical mode

The -g command line option of ovitos starts a graphical program session and the script will be run in the context of OVITO’s main window. This allows you to follow your script’s actions as they are being executed. This is useful for debugging purposes if you want to visually check the outcome of your script’s action during the development phase. Keep in mind that the viewports will only show pipelines that have been inserted into the current scene. Thus, it is necessary to explicitly call Pipeline.add_to_scene() to make your imported data visible in this mode.

Number of CPU cores

OVITO uses all available processor cores by default to perform some computations. To explicitly restrict the program to a certain maximum number of parallel threads, use the --nthreads command line parameter, e.g. ovitos --nthreads 1

Installing third-party Python modules in ovitos

OVITO’s embedded interpreter is a preconfigured version of the standard CPython interpreter with the ovito Python package built in. This makes it possible to run scripts both within the graphical program OVITO as well as through the ovitos command line interpreter. However, OVITO’s Python interpreter only ships with the NumPy and matplotlib preinstalled packages.

If you need to call other third-party Python packages from your OVITO scripts, it may be possible to install them in the ovitos interpreter using the normal pip or setuptools mechanisms (e.g., run ovitos -m pip install <package> to install a module from PyPI).

Installing Python extensions that include native code may fail, however, because such extensions may not be compatible with the build-time configuration of the embedded CPython interpreter. In this case, it is recommend that you instead install the ovito module in your system’s Python interpreter. Further instructions can be found here.