Back to News
quantum-computing

The QuSquare suite helped measure Greenberger Horne Zeilinger states on nine quantum processors

Ivy Delaney
Loading...
7 min read
0 likes
⚡ Quantum Brief
Researchers from the University of Deusto, Mondragon Unibertsitatea and other institutions in the Basque Country have benchmarked nine commercially accessible quantum processors using a common standard. The work demonstrates a new level of cross-platform evaluation, employing the QuSquare suite to measure the fidelity of Greenberger-Horne-Zeilinger (GHZ) states on devices from five vendors. Results reveal a significant disparity between the number of physical qubits and the size of GHZ states successfully generated; the best superconducting processors reached 27 and 24 qubits, less than a quarter of their total registers, while a 12-qubit trapped-ion processor passed the benchmark with its full register.
AI Audio Summary
0:00 / 0:00
Click to play
page-073-object-078.webp
Quantum News · Media Library

Researchers from the University of Deusto, Mondragon Unibertsitatea and other institutions in the Basque Country have benchmarked nine commercially accessible quantum processors using a common standard. The work demonstrates a new level of cross-platform evaluation, employing the QuSquare suite to measure the fidelity of Greenberger-Horne-Zeilinger (GHZ) states on devices from five vendors. Results reveal a significant disparity between the number of physical qubits and the size of GHZ states successfully generated; the best superconducting processors reached 27 and 24 qubits, less than a quarter of their total registers, while a 12-qubit trapped-ion processor passed the benchmark with its full register. The findings provide a baseline for future comparisons of quantum processor performance. QuSquare Suite Measures GHZ States Across Nine Processors The QuSquare suite facilitated the evaluation of multipartite entanglement across a diverse set of quantum processors, enabling a comparative analysis of their performance. Experiments used cloud access to nine processors, spanning superconducting and trapped-ion technologies, to benchmark generation of Greenberger-Horne-Zeilinger (GHZ) states, a key indicator of quantum system complexity. This cross-platform approach moves beyond isolated processor tests to provide a broader understanding of current quantum hardware capabilities. Analysis revealed a disconnect between calibration data and actual benchmark outcomes. Median calibration data proved poor predictors of performance on the GHZ benchmark. The statistical method used to determine success also significantly impacted results; requiring certified genuine multipartite entanglement (GME) reduced the number of processors successfully passing the benchmark from nine to four. These results provide a baseline for future cross-platform benchmarking of quantum processors, and the team intends to continue refining the methodology and expanding the scope of the evaluation. Benchmarking GHZ States Reveals Gap in Qubit Utilization GHZ state fidelity estimation, using direct fidelity estimation with a sample count independent of qubit number, revealed a performance disparity across evaluated quantum processors. This methodology, employed across nine commercially accessible processors, intentionally excluded error mitigation techniques to reflect typical end-user experiences. The structural properties of GHZ states, specifically their efficient preparation via Clifford gates, make them uniquely suited for benchmarking diverse quantum hardware. These states, requiring only Hadamard and controlled-NOT gates, circumvent limitations imposed by native gate sets or qubit technologies. This standardization allows for a more equitable comparison of performance across superconducting and trapped-ion platforms, revealing inherent strengths and weaknesses in each architecture. The largest GHZ states successfully generated ranged from three qubits on the Qmio processor to 27 qubits on IBM’s Nighthawk r2, demonstrating a wide variation in performance.

Superconducting Processors Achieve N^* of 27 and 24 Qubits Experiments conducted via remote cloud access reveal a stark contrast in multipartite entanglement generation between superconducting and trapped-ion quantum processors, with the best superconducting systems achieving N^* = 27 and N^* = 24 qubits. The results demonstrate that while superconducting processors boast higher qubit counts, their ability to maintain entanglement across a substantial portion of their register lags behind that of trapped-ion technology. The study evaluated nine commercially accessible processors from IBM Quantum, IQM, Rigetti, including the Qmio system operated by the Galicia Supercomputing Center. The benchmark focused on identifying the largest GHZ state whose estimated fidelity met a pre-defined threshold for genuine multipartite entanglement, a critical measure of quantum system coherence. This approach allowed for a direct comparison of performance across diverse architectures, highlighting the strengths and limitations of each platform. The IQM processors, with 20 and 54 qubits, achieved GHZ states comprising 65% and 37% of their registers, respectively, demonstrating a wider range of performance within the superconducting category. The work’s implementation of the QuSquare suite across these diverse platforms establishes a standardized approach for evaluating quantum processor capabilities and will likely be a valuable resource for the broader quantum computing community. Trapped-Ion Processors Demonstrate Full Register GHZ State Fidelity Experiments performed via remote cloud access reveal a distinct advantage for trapped-ion processors in generating high-fidelity Greenberger-Horne-Zeilinger (GHZ) states across their entire register size, a feat not yet matched by larger superconducting systems evaluated in the same study. While superconducting processors from IBM Quantum and IQM achieved the largest absolute qubit scales meeting the benchmark threshold, reaching a maximum of 27 qubits, these GHZ states comprised less than a quarter of each processor’s total qubit count. Requiring certified GME reduces the number of devices that succeed at any tested size from nine to four. GHZ State Preparation Probes Multi-Qubit Processor Capabilities GHZ state preparation circuits use a structure readily adaptable to diverse quantum processors, relying primarily on Hadamard and controlled-NOT gates, regardless of the underlying qubit technology or native gate set. This inherent flexibility allows for optimized circuit construction on virtually any digital quantum processor, a key feature for standardized benchmarking efforts. The selection of the most efficient circuit for a given processor is determined by minimizing the number of native two-qubit gates required after transpilation, streamlining the process and reducing potential error accumulation. All experiments were performed through remote cloud-access services, with the default compilation of each provider and without error mitigation, reflecting the performance experienced by end users. The multipartite entanglement benchmark, central to this work, assesses a quantum processor’s capacity to generate genuine multipartite entanglement through GHZ state preparation. The benchmark establishes a success criterion based on the largest GHZ state achievable while maintaining a specified fidelity level, effectively quantifying the processor’s multi-qubit capabilities. A binary search algorithm is employed to determine this maximum size, systematically testing progressively larger GHZ states until the fidelity falls below the acceptable threshold. This method provides a robust and efficient means of evaluating processor performance, pinpointing the limits of entanglement generation. Stabilizer measurements form the core of the fidelity estimation process, using randomly sampled operators to characterize the prepared GHZ state. “As a stabilizer state, |GHZN⟩ admits efficient preparation circuits composed exclusively of Clifford gates,” the researchers note, highlighting the practical advantages of using GHZ states for benchmarking. Several experimental studies have already demonstrated GHZ state preparation across various quantum platforms, including superconducting, trapped-ion, and silicon-based systems. Hu et al. 6817(1) on a 31-qubit superconducting processor, successfully certifying genuine multipartite entanglement using an entanglement witness. Edlbauer et al. further demonstrated the use of GHZ states to benchmark the all-to-all connectivity of an 11-qubit silicon atom processor, achieving genuine multipartite entanglement among up to eight nuclear spins. Peev and Sivkov recently evaluated GHZ-state preparation with up to 10 qubits on IBM’s 127-qubit ibm_brisbane processor, expanding the scale of these investigations. Cross-Platform Benchmarking Establishes Baseline for Future Studies Nine quantum processors, spanning superconducting and trapped-ion technologies, were subjected to a standardized assessment of multipartite entanglement generation, revealing discrepancies in performance despite comparable qubit counts. The benchmark used the QuSquare suite. This work moved beyond isolated demonstrations on single platforms, instead focusing on direct cross-platform comparison using a consistent methodology and metric. This highlights the sensitivity of entanglement benchmarks to statistical thresholds and the need for standardized evaluation protocols.

The team’s approach allowed for a quantitative assessment of GHZ state fidelity, establishing a baseline for future comparisons. The researchers emphasize the importance of standardized benchmarking in the rapidly evolving quantum computing landscape. The work’s focus on cloud-based access and the use of a common software suite, QuSquare, facilitates broader participation and reproducibility of these benchmarking experiments. This approach is essential for encouraging collaboration and accelerating progress in the field, as it allows researchers to evaluate processors remotely without requiring direct access to the hardware. 👉 More information🗞 Cross-Platform Benchmarking of GHZ-State Generation in Quantum Processors via Cloud Services✍️ Sergio Hernández et al.🧠 ArXiv: https://arxiv.org/pdf/2611.14114v1 More like thisQuantum AlgorithmsGoogle Quantum AI demonstrates contextuality at scaleQuantum Computing NewsInfleqtion’s Sqale entangles 30 logical qubits with 80 AtomsQuantum Research NewsResearchers digitize quantum Hamiltonians using dual-species atomsQuantum HardwareLiquid metals pass the test for linking quantum modules at 15 millikelvinStay currentSee today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals. Tags:

Read Original

Tags

quantum-hardware

Source Information

Source: Quantum Zeitgeist

Discussion

0 professional contributions

Sign in to join this professional discussion.

Be the first to add a constructive contribution.