Reproduce “MQT Bench: Benchmarking Software and Design Automation Tools for Quantum Computing” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Nils Quetschlich, Lukas Burgholzer, Robert Wille
MQT Bench offers a unified, cross-level benchmark suite for assessing quantum software and design automation tools, encompassing over 70,000 circuits ranging from 2 to 130 qubits across four abstraction levels. This resource supports evaluations for multiple compilers and hardware architectures, providing diverse circuit characteristics for rigorous empirical analysis.
Key insights
- A single, unified benchmark suite can effectively bridge the gap between high-level algorithmic descriptions and hardware-specific implementations by providing relevant circuits across all abstraction levels.
- Standardized and accessible benchmarking is crucial for improving the rigor of quantum software evaluations, fostering comparability among tools, and accelerating their development and optimization.
- The diversity of benchmark circuits, characterized by structural and operational properties, is essential for ensuring that software tools are evaluated generally and not overfit to specific problem instances.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.