Supported Platforms
Topics covered in this chapter:
Operating System and Hardware
The following are the supported platform, operating system, and processor combinations for FICO Xpress (unless stated otherwise in the following sections for specific components):
| Platform | Operating System | Processor | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
||||||||
|
|
|
||||||||
|
|
|
||||||||
|
|
|
Xpress Solver - Xpress Optimizer, Xpress NonLinear and Xpress Global
Xpress Solver is the framework that supports all of FICO's optimization solver technologies, including Xpress Optimizer, Xpress NonLinear and Xpress Global.
These can be used within our modeling language Xpress Mosel or directly from Python.
On Linux platforms, the minimum required version of the glibc library is 2.28.
Please refer to the installation instructions in Section Prerequisites for Linux Installations for dependencies needed for Linux distributions for running the Console Optimizer.
The use of key-based licences (see Section Using Key-based (Static or Web Floating) Licenses) requires a C++ runtime distribution, the minimum required version is 14. In the absence of a suitable C++ version, the file-based licensing method needs to be used.
Python Support
Xpress Solver can be used with the following Python versions.
| Platform | Python Support | ||||||||
|---|---|---|---|---|---|---|---|---|---|
|
|
R support
The R interface for Xpress Optimizer can be used with R versions 4.0 and newer on macOS and Linux. The precompiled R-package for Windows supports R version 4.0.
Xpress Insight Compute Interface support
The Xpress Optimizer libraries can be configured to interact with a remote Insight server. This is supported on all platforms listed above (Section Operating System and Hardware).
In addition, access to an Xpress Insight 5 server is required (see the Insight 5 Installation Guide, and in particular the chapter Supported Platforms).
GPU support for PDHG (beta release)
The primal-dual hybrid gradient (PDHG) linear optimization solver using a NVIDIA® CUDA®-capable GPU has been tested on the following operating systems and platforms:
| Platform | Operating System | |||||
|---|---|---|---|---|---|---|
|
|
|||||
|
|
|||||
|
|
Hardware requirements:
-
NVIDIA GPU with microarchitecture Turing, Ampere, Ada Lovelace, Hopper or Blackwell.
Please refer to the installation instructions in Section GPU installation guidelines for PDHG for additional software and driver requirements.
Xpress Mosel
The Xpress Mosel language allows the user to define his models in a form that is close to algebraic notation and to solve them in the same environment.
It interfaces to statistics packages such as R or Matlab. Mosel provides a module python3.dso that implements functionality for exchanging data between a Mosel model and Python 3 (C Python) and for calling Python 3 scripts.
- Xpress Workbench is the premier choice as IDE (Integrated Development Environment) for standalone Mosel models and Xpress Insight apps. It integrates with Insight for remote debugging. Xpress Workbench is available for Windows and macOS as part of the Xpress installer, and as part of Docker images.
Python Support
Mosel provides a module python3.dso that implements functionality for exchanging data between a Mosel model and Python 3 (CPython) and for calling Python 3 scripts. The Mosel run-time library loads and runs the Python interpreter. Xpress Mosel can be used with the following Python versions:
- Python 3.10 to Python 3.14
The Python package moselpy for embedding Mosel models into Python programs supports the same Python versions as the Mosel module.
R support
The R interface for Xpress Mosel makes it possible to easily exchange data with R and execute R scripts, or evaluate expressions in the R language, from within a Mosel model.
- Mosel supports R versions 4.0 to 4.1.x
Note: Download the version of R from the R Project web site at www.r-project.org, targeting the same platform as Mosel. Therefore, if you have Windows 64-bit Mosel installed, download a matching R version.
Data Sources
Mosel can connect to data in memory through files, databases and web services. It connects to any ODBC-enabled data source, has a specific Oracle driver and drivers for Excel, CSV, XML, JSON as well as its own DAT format. Users can implement custom drivers and use Mosel's free form reading and writing capabilities.
Xpress Solver - Xpress Kalis
Xpress Kalis provides access to the Artelys Kalis Constraint Programming (CP) solver from a Mosel module allowing the user to formulate and solve CP models in the Mosel language.
Xpress Kalis combines a finite domain solver and a solver over continuous (floating point) variables. All data handling facilities of the Mosel environment, including data transfer in memory and ODBC access to databases can be used with Xpress Kalis.
| Platform | Operating System | Processor | ||||||
|---|---|---|---|---|---|---|---|---|
|
|
|
||||||
|
|
|
Xpress Solver - Knitro
Xpress Knitro is a nonlinear solver provided by Artelys. It is accessible through Xpress Solver and is supported on the same platforms (see Section Operating System and Hardware).
Interfaces
Java
The interfaces for calling Xpress libraries from Java require a minimum of Java 8 (internal version: 1.8.0); compatible distributions of Oracle Java and OpenJDK are supported.
The Mosel module mosjvm works with Java 8, 11 and 17. Java 8 is not available on macOS ARM. The module is tested on Amazon Corretto.
.NET
The interfaces for calling Xpress Solver and Mosel libraries from .NET are available for Windows and Linux platforms, and require .NET 8.
C++
The interface for calling Xpress Solver from C++ requires a minimum of C++ 17.
© 2001-2026 Fair Isaac Corporation. All rights reserved. This documentation is the property of Fair Isaac Corporation (“FICO”). Receipt or possession of this documentation does not convey rights to disclose, reproduce, make derivative works, use, or allow others to use it except solely for internal evaluation purposes to determine whether to purchase a license to the software described in this documentation, or as otherwise set forth in a written software license agreement between you and FICO (or a FICO affiliate). Use of this documentation and the software described in it must conform strictly to the foregoing permitted uses, and no other use is permitted.