Skip to content

Combinatorial Optimization

The Classiq combinatorial optimization platform is a quantum software engine for optimization problems that you define. The engine tackles your formulated real-world optimization challenges and generates customized quantum programs. You choose whether to run the programs on a quantum backend or a classical simulation, and receive the optimized solution. To gain further insights regarding the synthesis process, you can examine the synthesized program with the Classiq analyzer module.

Formulating Problems

You describe new optimization problems using the Python SDK package. The problem is formulated using PYOMO, a Python-based, open-source optimization modeling language. The language supports a wide variety of problem types, such as integer linear programming, quadratic programming, graph theory problems, and SAT problems. Read in-depth reviews of the language’s capabilities in [1] and [2] [3] . The basics of problem modelling in PYOMO and a complete example are in the problem formulation section.

The Classiq platform supports an extensive set of modeling configurations for your use (supported modeling). See examples of graph problems and integer linear programming in the problem library.

Solving Optimization Problems

The core Classiq capabilities are generation of a designated quantum solution, and execution of the generated algorithm on a quantum backend. Both capabilities are available as part of the Python SDK and demonstrated in the problem solving section.

The Classiq platform relies on the QAOA penalty algorithm to solve optimization problems.

The solving algorithms are flexible and customizable, as demonstrated in the solver customization section.


[1] Pyomo Documentation 6.0.1,

[2] Prof. Jeffrey Kantor’s Pyomo Cookbook

[3] J. D. Siirola, Introduction to Pyomo: The optimization foundation for IDAES, (2018).