Researchers from ParityQC have demonstrated a 52-qubit quantum Fourier transform (QFT) on an IBM Quantum Heron r3 processor, establishing a new performance record for the size of such a circuit and nearly doubling a previous benchmark achieved on trapped-ion hardware. Scaling QFTs is particularly difficult due to the challenges of maintaining performance as circuit depth and noise accumulate; however, the ParityQC team employed a novel parity-based construction method to eliminate the need for explicit qubit routing. “It’s not about the qubit count,” said ParityQC co-founder and co-CEO Wolfgang Lechner, “With our method, we were actually able to reduce the errors and still achieve this doubling.” This advance, executed on IBM Quantum hardware, suggests that algorithmic improvements can deliver substantial gains even with current quantum technology, potentially accelerating progress toward practical applications in fields like optimization and quantum chemistry.
Qubit Quantum Fourier Transform Achieved on IBM Heron Processor
The successful execution of a 52-qubit Quantum Fourier Transform (QFT) by ParityQC on IBM Quantum hardware marks a significant step forward in quantum computation, demonstrating that algorithmic innovation can outpace simply adding qubits. The QFT, a cornerstone of many quantum algorithms, is notoriously difficult to scale due to the challenges of maintaining coherence and managing errors as circuit complexity increases. ParityQC addressed these limitations with a novel method called Parity Twine, which fundamentally alters how quantum information is represented and propagated.
Instead of relying on SWAP operations to route qubits for interaction, a process that adds circuit depth and amplifies errors, the team tracked parity information, effectively merging the logistics of routing with the algorithm itself. “Previously, it was like you do the swapping, which is the logistics, and then the algorithm,” explained ParityQC co-founder and co-CEO Wolfgang Lechner. “In our case, we are able to merge these two things.” This approach allows for the transfer of information using CNOT gates, reducing the overall gate count and circuit depth. The conceptual shift involves delocalizing quantum information across multiple qubits, creating overlapping pathways that perform computations and propagate correlations simultaneously. Lechner elaborated that this allows multi-qubit interactions, typically resource-intensive, to be potentially reduced to simpler operations. The team’s results, benchmarked against circuits produced using the Qiskit transpiler, demonstrate improved process fidelity, a key metric for evaluating the accuracy of quantum algorithms, particularly as system size increases.
Parity Twine Method Eliminates SWAP-Based Routing Overhead
The pursuit of scalable quantum computation currently hinges on overcoming limitations imposed by the architecture of existing hardware and the algorithms designed to run on it. While increasing qubit counts remains a primary focus, recent advances demonstrate that algorithmic innovation can yield substantial performance gains even on near-term devices. This achievement is particularly notable as it nearly doubles a previous QFT benchmark established on trapped-ion hardware. Traditional superconducting quantum processors rely on SWAP operations to facilitate interactions between non-adjacent qubits, a process that introduces additional gates, increases circuit complexity, and amplifies accumulated errors. ParityQC’s approach, termed Parity Twine, fundamentally alters this paradigm by shifting the focus from individual qubit states to relationships between qubits. Instead of physically swapping quantum states, the team transfers parity information, the correlations between qubits, using sequences of CNOT gates.
Lechner further clarified that the result is a reduction in gate count and circuit depth, enabling more complex computations with improved fidelity. The team’s experiments, conducted on the IBM Quantum Heron r3 processor, demonstrated improved process fidelity compared to circuits generated using the Qiskit transpiler, particularly as the system size increased. “Our experiments show that this is the best hardware around at the moment,” Lechner stated, emphasizing the importance of a collaborative quantum ecosystem where algorithmic advancements and hardware improvements mutually reinforce each other. This method doesn’t just improve the QFT; it offers a pathway to more efficient implementations of algorithms used in optimization, simulation, and quantum chemistry, suggesting a broader impact on the field.
It’s not about the qubit count,” said ParityQC co-founder and co-CEO Wolfgang Lechner, who noted that the new work nearly doubles a previous QFT benchmark set in on trapped-ion hardware.
Process Fidelity Benchmarks Demonstrate Algorithm Performance Gains
ParityQC is pushing the boundaries of quantum computation not simply by increasing qubit counts, but by fundamentally altering how quantum information is processed. The team’s success hinges on a method called Parity Twine, which addresses the persistent challenge of routing qubits, connecting them for interaction, on superconducting processors. Traditional approaches rely on SWAP operations to move quantum states, adding circuit depth and accumulating errors. Parity Twine circumvents the need for many SWAP gates by tracking parity information, the relationships between qubits, rather than the state of individual qubits. This allows for a more streamlined computation, reducing gate count and circuit depth.
The conceptual shift involves a delocalization of quantum information, allowing interactions to be reframed as single-qubit operations. “In a standard algorithm, you would just have a yellow qubit that goes across the chip via SWAP gates and is somewhat localized,” Lechner explained, contrasting this with the parity method where “the green and the yellow sort of share one qubit and the colors start to mix.” This approach isn’t merely about optimizing existing circuits; it’s about redefining how quantum computations are represented at a higher level. To validate their method, ParityQC employed process fidelity as a key metric, measuring the accuracy of the entire QFT algorithm. They compared Parity Twine against highly optimized circuits generated by the Qiskit transpiler, finding improved performance, especially as the system size increased. Lechner asserts that this metric is crucial for comparing different quantum devices, regardless of their underlying technology. The results highlight the power of a collaborative quantum ecosystem, where platform tools like Qiskit provide a foundation for innovation from startups and academic institutions alike.
Even if you think about addition-adding two numbers-if you do this on a quantum computer, it’s based on QFT,” Lechner said.
QFT as a Versatile Benchmark for Quantum Hardware Performance
The pursuit of scalable quantum computation increasingly relies on standardized benchmarks to assess progress, and the quantum Fourier transform (QFT) is rapidly emerging as a particularly versatile tool for evaluating hardware performance. Beyond its fundamental role within algorithms like Shor’s and quantum phase estimation, the QFT’s sensitivity to hardware limitations, particularly qubit connectivity and accumulated noise, makes it an ideal stress test for current systems. This achievement is significant not simply for the qubit count, but for the methodology employed. This shift allows for a merging of algorithmic design and hardware routing, streamlining the computation.
The team’s process fidelity metric, designed to capture the combined effects of circuit depth, noise, and compilation strategy, allows for meaningful comparisons between different quantum hardware platforms, regardless of their underlying technology, superconducting qubits, ions, or atoms. “We want to compare all these different quantum devices,” Lechner said. “This shouldn’t matter. We want to compare them, and the process fidelity is one way to do that.” Ultimately, the QFT’s widespread applicability as a building block for numerous algorithms solidifies its position as a standard component for evaluating and advancing quantum device capabilities, and a crucial indicator of progress toward practical quantum applications.
With our method, we were actually able to reduce the errors and still get this doubling.
