Quadratic Unconstrained Binary Optimization (QUBO) is a versatile framework for modeling combinatorial optimization problems. This study benchmarks five software-based QUBO solvers: Neal, PyTorch (CPU), PyTorch (GPU), JAX, and SciPy, on randomly generated QUBO matrices ranging from 1000x1000 to 45000x45000, under six convergence thresholds from 10^-1 to 10^-6. We evaluate their performance in terms of solution quality (energy) and computational time. Among the solvers tested, Neal achieved the lowest energy values but was limited to problems with up to 6000 variables due to high memory consumption. PyTorch produced slightly higher energy results than Neal but demonstrated superior scalability, solving instances with up to 45000 variables. Its support for GPU acceleration and CPU multi-threading also resulted in significantly shorter runtimes. JAX yielded energy values slightly above those of PyTorch and was limited to 25000 variables, with runtimes comparable to PyTorch on GPU. SciPy was the most constrained solver, handling only up to 6000 variables and consistently producing the highest energy values with the longest computation times. These findings highlight trade-offs between solution quality, scalability, and runtime efficiency, and suggest that PyTorch is the most balanced choice for large-scale QUBO problems when computational resources permit.
Quadratic Unconstrained Binary Optimization (QUBO) is a versatile framework for modeling combinatorial optimization problems. This study benchmarks five software-based QUBO solvers: Neal, PyTorch (CPU), PyTorch (GPU), JAX, and SciPy, on randomly generated QUBO matrices ranging from 1000x1000 to 45000x45000, under six convergence thresholds from 10^-1 to 10^-6. We evaluate their performance in terms of solution quality (energy) and computational time. Among the solvers tested, Neal achieved the lowest energy values but was limited to problems with up to 6000 variables due to high memory consumption. PyTorch produced slightly higher energy results than Neal but demonstrated superior scalability, solving instances with up to 45000 variables. Its support for GPU acceleration and CPU multi-threading also resulted in significantly shorter runtimes. JAX yielded energy values slightly above those of PyTorch and was limited to 25000 variables, with runtimes comparable to PyTorch on GPU. SciPy was the most constrained solver, handling only up to 6000 variables and consistently producing the highest energy values with the longest computation times. These findings highlight trade-offs between solution quality, scalability, and runtime efficiency, and suggest that PyTorch is the most balanced choice for large-scale QUBO problems when computational resources permit.
Quadratic Unconstrained Binary Optimization (QUBO) provides a versatile framework for representing NP-hard combinatorial problems, yet existing solvers often face trade-offs among speed, accuracy, and scalability. In this work, we introduce a quantum-inspired solver (QIS) that unites branch-and-bound pruning, continuous gradient-descent refinement, and quantum-inspired heuristics within a fully adaptive control architecture. We benchmark QIS3 against eight state-of-the-art solvers, including genetic algorithms, coherent Ising machines, simulated bifurcation, parallel tempering, simulated annealing, our prior QIS2 version, D-Wave's simulated-annealing (Neal), and Gurobi on three canonical QUBO problem classes: Max-Cut, NAE-3SAT, and Sherrington-Kirkpatrick spin glass problems. Under a uniform runtime budget, QIS3 attains the best solution on nearly all instances, achieving optimality in 94% of max-cut instances. These results establish QIS3 as a robust, high-performance solver that bridges classical exact strategies and quantum-inspired heuristics for scalable QUBO optimization.
The Quadratic Unconstrained Binary Optimization (QUBO) model is widely used to solve combinatorial optimization problems by determining the optimal assignment of binary variables in a quadratic formulation. Given its versatility, extensive research has been conducted on developing QUBO solvers and converting combinatorial optimization problems into equivalent QUBO formulations. This paper introduces QUBO++, a novel C++ symbolic mathematical computation library specifically designed for QUBO modeling. It provides a comprehensive set of operators and functions for symbolic computation, enabling seamless transformation of optimization problems into QUBO models. While several Python-based libraries exist for similar purposes, QUBO++ is the only dedicated C++ library that supports multithreaded acceleration via Intel Threading Building Blocks (oneTBB). In addition to its modeling capabilities, QUBO++ includes two fundamental QUBO solvers: the Exhaustive Solver, which is useful for verifying small-scale problems, and the Easy Solver, designed for efficient testing and debugging. Compared to existing Python-based QUBO modeling tools such as PyQUBO, Fixstars Amplify, and Tytan, QUBO++ demonstrates significantly higher performance, achieving speedups of over 1,700 times in certain scenarios. By leveraging QUBO-based problem-solving techniques, QUBO++ serves as a powerful and versatile tool for addressing complex optimization challenges in both industrial and societal applications.
Recent advancements in quantum computing have demonstrated significant potential for solving combinatorial optimization problems, like the quadratic knapsack problem, a constrained binary optimization problem. How- ever, current quantum and quantum-inspired algorithms often require transforming these constrained problems into an unconstrained form, known as Quadratic Unconstrained Binary Optimization (QUBO). Such transfor- mations can significantly impact the algorithms’ speed and efficiency. In this study, we evaluate five existing transformation methods and propose four novel approaches. We assess all nine methods using Simulated An- nealing and find that three of our approaches outperform existing methods in terms of execution time and the quality and quantity of feasible solutions found. Additionally, we tested these transformations on quantum annealers, which were unable to solve even small problem instances, due to limitations in connectivity and error rates. However, our results highlight the advantages of the new approaches, which reduce the total num- ber of variables in the QUBO representation. This is a critical factor for enhanced performance on emerging quantum hardware, since it also reduces the required number of qubits and the embedding chain lengths.
In this paper, we propose a hybrid framework to solve large-scale permutation-based combinatorial problems effectively using a high-performance quadratic unconstrained binary optimization (QUBO) solver. To do so, transformations are required to change a constrained optimization model to an unconstrained model that involves parameter tuning. We propose techniques to overcome the challenges in using a QUBO solver that typically comes with limited numbers of bits. First, to smooth the energy landscape, we reduce the magnitudes of the input without compromising optimality. We propose a machine learning approach to tune the parameters for good performance effectively. To handle possible infeasibility, we introduce a polynomial-time projection algorithm. Finally, to solve large-scale problems, we introduce a divide-and-conquer approach that calls the QUBO solver repeatedly on small sub-problems. We tested our approach on provably hard Euclidean Traveling Salesman (E-TSP) instances and Flow Shop Problem (FSP). Optimality gap that is less than 10% and 11% are obtained respectively compared to the best-known approach.
The calibration of suspension geometry involves highly nonlinear kinematic relationships and leads to challenging optimization landscapes that are difficult to explore efficiently with classical local methods. Quadratic Unconstrained Binary Optimization (QUBO) provides a unified discrete formulation that enables the use of a wide range of metaheuristic solvers, but its practical behavior in realistic engineering-inspired problems remains insufficiently benchmarked. This paper presents a computational benchmarking study of classical, parallel, and hybrid metaheuristic solvers applied to a QUBO-formulated double wishbone suspension geometry problem. A symbolic multi-body kinematic model is constructed and discretized into a large-scale QUBO instance capturing camber and caster tracking objectives across multiple roll conditions. Using a fixed low-resolution binary encoding, we systematically evaluate solver performance in terms of objective value, runtime, and time-to-solution trade-offs. The benchmark includes standard simulated annealing and tabu search, parallel simulated annealing, population-based annealing, bandit-controlled hybrid heuristics, and continuous-relaxation-based ADMM methods with and without annealing refinement. Extensive experiments conducted on a Euro-HPC pre-exascale system demonstrate that parallel and hybrid solvers can achieve substantial reductions in wall-clock time—often exceeding two orders of magnitude—while attaining objective values comparable to classical simulated annealing. The results reveal clear trade-offs between solution quality and computational efficiency, and highlight how solver structure influences performance on large QUBO instances derived from symbolic engineering models. Rather than focusing on final design optimality, this study provides a reproducible reference benchmark and practical insights into solver behavior for QUBO-based engineering optimization problems.
We introduce VeloxQ, a fast solver for Quadratic Unconstrained Binary Optimization (QUBO) problems, which are central to many real-world optimization tasks. Unlike approaches that depend on emerging quantum hardware, VeloxQ can be deployed on conventional computing infrastructure. We benchmark VeloxQ against state-of-the-art QUBO solvers from several families. These include quantum annealers, specifically D-Wave's Advantage and Advantage2 platforms; the digital-quantum BF-DCQO algorithm for Higher-Order Unconstrained Binary Optimization (HUBO) developed by Kipu Quantum; physics-inspired algorithms including Simulated Bifurcation, Parallel Annealing, and tropical tensor networks; and conventional methods including CPLEX, brute force, BEIT's Chimera solver, and Branch-and-Bound variants. The benchmark suite covers native quantum-annealer topologies, embedded all-to-all instances, HUBO-derived instances, planted-solution instances, certified-solver regimes, and dense Branch-and-Bound test cases. Across the benchmark suite, VeloxQ delivers competitive solution quality and runtime, and in several regimes outperforms the compared solvers. VeloxQ also demonstrates strong scalability. Among the solvers considered in this study, it was the only method we could run on the largest sparse instances within our computational budget, including problems with up to 108 sparsely connected variables. These findings position VeloxQ as a competitive and practical tool for tackling large-scale QUBO/HUBO problems, offering a practical alternative to existing quantum and classical optimization methods.