Home > Products > MOSEK


The MOSEK optimization software is designed to solve large-scale mathematical optimization problems.  Below we provide an overview of MOSEK and If you have additional questions about our products feel free to contact us or browse the online documentation.

Problem types MOSEK can solve:

  • Linear problems.
  • Conic quadratic problems.
  • Semidefinite problems. (Positive semidefinite matrix variables).
  • Quadratic and quadratically constrained problems.
  • General convex nonlinear problems. 
  • Mixed integer linear, conic and quadratic problems.

Technical highlights of MOSEK are:

  • Problem size is only limited by the available memory.
  • An interior-point optimizer with basis identification.
  • Both primal and dual simplex optimizers for linear programming.
  • Highly efficient presolve  for reducing problem size before optimization.
  • A special optimizer is available for problems with network flow structure.
  • For mixed integer problems MOSEK implements a branch & bound & cut algorithm.

Strengths of MOSEK:

The strong point of MOSEK is the interior-point point optimizer for continous linear, quadratic and conic problems. The interior-point optimizer is simply state-of-the-art.

Parallel capabilities:

  • The interior-point optimizer is capable of exploiting multiple CPUs/cores.
  • A concurrent optimizer is available that makes it possible to solve one problem with different optimizers simultaneously.
  • The default mixed integer optimizer is parallelized and is run-to-run deterministic.

Other features of MOSEK:

  • Reads and writes industry standard formats such as the MPS and LP formats.
  • Includes tools for infeasibility diagnosis and repair.
  • Sensitivity analysis for linear problems.


The MOSEK optimizer includes several interfaces.

The most basic interface called optimizer API is a highly efficient matrix oriented interface that is direct interface to optimizers available in MOSEK. The optimizer API is available for

  • C/C++,
  • Java,
  • .NET,
  • and Python.


MOSEK also provides an object orientated interface called Fusion for building conic optimization problems in a small amount of time. Fusion deals directly with variables and constraints and hence make it possible to express an optimization models very similar to the way it is formulated. Fusion is available for

  • Java,
  • MATLAB, 
  • .NET,
  • and Python.


Finally MOSEK comes with an optimization toolbox for MATLAB and it works with various third party tools and products such as R.



We have over a decade of experience providing support to our worldwide customer base.