Quantinuum and Sandia National Laboratories have launched the Quantum Universal Operations Performance System, or QUOPS, a new benchmark designed to move beyond qubit counts and assess what quantum computers can actually do. QUOPS measures performance across physical and logical qubits using two metrics, Q, representing circuit size, and Ω, measuring operations per second, and is applicable to diverse quantum technologies.

The system focuses on identifying “the boundary of a system’s capability region,” providing a standardized way to track progress toward practical, fault-tolerant quantum computing as the field shifts from noisy intermediate-scale to application-scale systems. QUOPS aims to give buyers and researchers a common language for evaluating quantum systems and understanding the gap between current capabilities and the 10⁹-10¹² operations needed for complex workloads.

QUOPS Measures System Performance Beyond Traditional Metrics

The system, developed by Sandia National Laboratories with contributions from Quantinuum and NVIDIA, moves beyond assessing quantum computers by qubit count and gate fidelity, instead focusing on demonstrable computational success. This shift in evaluation is driven by the limitations of existing metrics as quantum computing transitions from Noisy Intermediate-Scale Quantum (NISQ) to Fault-Tolerant Application-Scale Quantum (FASQ) systems. Component-level measurements, while still essential for engineering improvements, no longer fully capture a system’s ability to execute complex workloads, particularly those requiring error correction.

“As organizations move from experimentation toward larger-scale and potentially on-premise quantum systems, they need to know a simple thing: What computation can a machine actually execute successfully?” the QUOPS documentation states, highlighting the need for a complete system assessment. The architecture-agnostic nature of QUOPS is a key feature, allowing for consistent benchmarking across diverse quantum technologies.

Unlike metrics tied to specific hardware modalities or quantum error correction codes, QUOPS applies the same randomized workloads to different systems, measuring the largest benchmark circuit size that passes a defined success threshold. This standardized approach addresses a significant challenge for potential buyers and governments seeking objective procurement criteria. “QUOPS provides a simpler system-level reference point: Q describes the size of computation a machine can execute, while Ω describes its effective throughput,” the documentation explains.

Quantinuum, a publicly listed company trading on Nasdaq as QNT with about 630 people and $3.11 billion in total funding, has been actively pursuing fault-tolerant quantum computation with its trapped-ion technology, recently launching a 98-qubit Helios system. The company received a grant of $100M from the U.S. Department of Commerce, CHIPS Act for trapped-ion manufacturing.

The company’s collaboration with High Performance Computing (HPC) leader HPE, expanding in 2026, underscores the importance of integrating quantum systems with existing computational infrastructure. Similarly, a partnership with Synopsys integrates quantum computing into engineering design workflows, using advanced quantum algorithms. Helios, with all-to-all connectivity inherent in trapped-ion qubits, achieves larger capability regions but at a slower operational rate. These commercial collaborations, alongside research efforts with institutions like Hiverge, which developed a multi-agent AI platform to autonomously discover quantum algorithms, demonstrate Quantinuum’s commitment to translating theoretical advances into practical applications, the company says.

The QUOPS benchmark, while new, has already been applied to several vendors’ hardware, revealing trade-offs between system characteristics. Figures released alongside the QUOPS launch show that superconducting systems like Willow and Boston, characterized by fast gate speeds but limited connectivity, exhibit smaller capability regions and lower Q values compared to Quantinuum’s Helios system.

This illustrates how QUOPS provides a two-dimensional view of performance, allowing for nuanced comparisons beyond single-number metrics. The system’s anti-gaming provisions are also intended to ensure fair comparisons, preventing vendors from optimizing solely for the benchmark rather than overall computational power, according to Quantinuum. By focusing on what a quantum system can actually deliver, QUOPS offers a more practical and relevant measure of performance than traditional metrics alone.

QUOPS Enables Objective Quantum System Procurement

QUOPS establishes a quantifiable link between quantum hardware and successful computation, reporting a “Q” value representing the largest benchmark circuit size a system can reliably execute. The system’s architecture-agnostic design allows for performance comparisons across diverse quantum modalities, including superconducting and trapped-ion systems, offering a standardized assessment previously unavailable to buyers and researchers, the company says.

The development of QUOPS addresses a critical need as quantum computing transitions from exploratory research toward practical applications requiring billions of operations and thousands of qubits. Traditional metrics provide limited insight into a system’s ability to handle complex workloads, particularly when factoring in the overhead of error correction and decoding.

As stated in the QUOPS documentation, the system measures “what a quantum system can actually execute successfully—including the effects of error correction, decoding, mitigation, compilation, and other system-level factors.” This complete approach is particularly relevant for high-performance computing centers evaluating the potential of quantum integration and customers seeking systems capable of tackling specific computational challenges.

“Willow and Boston have smaller capability regions and QUOPS scores but higher QUOPS rates; while Helios reaches larger capability regions and QUOPS scores but lower QUOPS rates,” the documentation notes, illustrating how QUOPS facilitates a more informed comparison than relying solely on individual hardware specifications. Quantinuum’s involvement in the development of QUOPS reflects the company’s commitment to advancing fault-tolerant quantum computing.

This expertise informed the design of QUOPS, ensuring its applicability to a broad range of architectures and fault tolerance levels, the company states. The company’s strategic partnerships, including collaborations with HPE and Mitsubishi Electric, further underscore its dedication to integrating quantum computing into existing computational infrastructure and industrial workflows. The system’s utility extends beyond simple benchmarking, offering a practical tool for procurement and planning.

HPC centers can use QUOPS to assess when quantum computing will become advantageous for real-world workloads, while customers can define specific performance thresholds in requests for proposals. “QUOPS can also provide a practical layer for quantum procurement and planning,” the documentation states, “Customers may have a goal of procuring a system that can, for example, run a trillion error-free operations.” By providing a common system-level yardstick, QUOPS aims to streamline the evaluation process and facilitate more objective decision-making.

These efforts, alongside a grant of $100M from U.S. Department of Commerce, CHIPS Act for trapped-ion manufacturing, highlight Quantinuum’s commitment to driving innovation across the entire quantum stack, by the company’s account. “As quantum computers become fault tolerant, success will no longer be defined simply by how many qubits a machine contains or how low its error rates are.

It will be defined by the computation the machine can deliver,” the documentation concludes, positioning QUOPS as a key tool for measuring that crucial capability and charting the course toward a more powerful and practical quantum future.

Helios, Willow, and Boston Compared Using QUOPS Benchmarks

The newly introduced Quantum Universal Operations Performance System (QUOPS) reveals distinct performance profiles among leading quantum processors, with Quantinuum’s Helios demonstrating a larger capability region compared to the superconducting systems Willow and Boston. While Willow and Boston exhibit faster operational rates due to their characterized fast gate speeds, Helios achieves greater computational scale through its all-to-all connectivity and integration of technologies like NVIDIA GB200 superchips for real-time error decoding, Quantinuum claims.

The benchmark’s architecture-agnostic design allows for comparative analysis across diverse modalities, as demonstrated by the initial results from Quantinuum, Google, and IBM processors. Specifically, the data shows Willow and Boston achieving higher QUOPS rates but constrained by smaller capability regions, while Helios extends the boundary of achievable computation, albeit at a lower operational rate. This trade-off between speed and scale is further influenced by error mitigation techniques, which introduce sampling overheads quantified by the QUOPS rate metric.

Quantinuum’s work with Hiverge on the “Hive” multi-agent AI platform, validated on the System Model H2, showcases an autonomous approach to discovering quantum algorithms. These efforts, combined with the QUOPS results, position Helios as a system capable of tackling increasingly complex computational challenges, even as Willow and Boston maintain an advantage in speed for certain workloads. The QUOPS framework also highlights the importance of connectivity in quantum processors.

This is not to suggest a simple superiority of one architecture over another, but rather a demonstration of how different design choices impact performance as measured by QUOPS. The system’s ability to quantify this trade-off is important for both researchers and potential buyers seeking to understand the strengths and weaknesses of competing quantum systems.

The data released alongside QUOPS demonstrates that all three systems can trade speed for larger circuits using error mitigation, a commonly assumed capability now precisely quantified by the QUOPS rate. Quantinuum’s R&D expansion into Singapore, establishing a center to host the Helios quantum computer, reflects a global strategy for advancing quantum technology. This move, supported by the Singapore Economic Development Board, positions Quantinuum as a key player in the international quantum landscape.

Additionally, the company’s partnership with Aramco to explore quantum computing for energy challenges and its collaboration with Rolls-Royce to advance fluid dynamics simulations demonstrate a focus on practical applications, Quantinuum says. These partnerships, alongside the development of a backpropagation algorithm for evaluating parameter gradients in quantum circuits, underscore Quantinuum’s complete approach to quantum computing, encompassing hardware, software, and algorithm development. The benchmark’s architecture-neutrality and anti-gaming provisions further enhance its value as a universal standard in a rapidly evolving field.

Source: https://www.quantinuum.com/blog/introducing-the-quantum-universal-operations-performance-system-quops

More like this