A new benchmark, called Quantum Universal Operation Performance System (QUOPS), has been developed by a team led by Sandia National Laboratories in the U.S. This benchmark provides a universal way to measure the computational power of quantum computers. Unlike traditional methods that focus on individual qubits and their error rates, QUOPS evaluates how large and complex a quantum circuit a computer can run successfully and how quickly it can perform these operations. This approach offers a more comprehensive view of a quantum computer's overall performance.
The QUOPS benchmark was tested on quantum hardware from major companies like Google, IBM, and Quantinuum. The results showed a wide range of performance levels, influenced by the design and construction of each system. Quantinuum’s Helios-1 trapped-ion processor achieved the highest QUOPS score of more than 1,500. This was due to its flexible layout, which allows qubits to communicate with each other more freely. However, the system is slower because moving the ions between qubits takes more time.
In comparison, superconducting processors from Google and IBM use a rigid grid layout for their qubits. This layout limits how qubits can interact, resulting in lower QUOPS scores—around 200. However, these systems are much faster in performing operations. For example, Google's Willow processor can handle 20 million operations per second, which is a significant advantage in speed.
The study highlights the current gap between the capabilities of quantum computers and the requirements needed to solve complex real-world problems. For instance, tasks such as breaking RSA 2048 encryption or modeling complex molecules for chemical research would require QUOPS scores in the range of 250 million to 340 million. The highest score currently achieved on physical hardware is only 1,824, showing a gap of about 100,000 times. To close this gap, researchers would need to significantly reduce error rates, increase the number of physical qubits, or develop more efficient quantum computer designs.
The research underscores the importance of having a single, transparent benchmark like QUOPS to track progress in quantum computing. The authors suggest that using rigorous benchmarking tools can help scientists and stakeholders independently verify and compare the computational capabilities of different quantum systems. This could lead to more consistent progress and better understanding of the field’s development.
New Benchmark Reveals Quantum Computers' Performance Gaps and Limitations
AI-rewritten from original reportingHow it works
quantumcomputingbenchmarkquopsqubitssandiaquantinuum
Original sources:
- 🇺🇸Phys.org



