For the first time, it's been shown that a large-scale IBM quantum computer can outdo other quantum technologies like annealers and trapped ion machines, Q-CTRL claims that the study proves that it is at the top of the field and could speed up the introduction of quantum computing.
Quantum optimization could greatly improve important areas like transportation, logistics, machine learning, and detecting financial fraud. McKinsey estimates that just in logistics, quantum computing could be worth $200-500 billion a year by 2035, but only if the quantum industry can solve these problems.
Past benchmarks showed that gate-based quantum computers didn't perform as well as hoped due to hardware errors. They often gave results that were no better than random guesses or only slightly better in certain situations.
Because of this, quantum annealers were seen as the better option for solving real-world optimisation problems. It was thought that today's quantum computers were far from being able to tackle these significant industry issues.
Q-CTRL has changed this view by tackling the error problem. They've combined new ways of running problems on the hardware with their software that manages performance on IBM's large-scale quantum computers. This approach has improved performance without needing to change the hardware.
Tests proved that Q-CTRL's technology made a quantum optimisation problem run on a 127-qubit IBM quantum computer up to 1,500 times more likely to give the right answer compared to an annealer.
Q-CTRL's breakthroughs with IBM's technology mean that quantum optimisation algorithms can now find the right solutions to tough problems more reliably and on a larger scale than before. This could outperform results from other quantum technologies and save costs because it takes fewer tries to get the right answer.
IBM Quantum VP Jay Gambetta said: "To make quantum computing useful, we need to build a quantum computer and find practical algorithms. IBM is working on the first part, and I'm excited to see Q-CTRL making progress on the second part by providing practical solutions to optimization problems that matter to businesses. Their software with an improved algorithm, along with our large-scale quantum computers, shows that we're closer to a quantum advantage than we thought."
Q-CTRL Founder Michael J. Biercuk said: "It's been hard for anyone, even experts, to fully use large-scale machines. Our new results show what's now possible with current technology, clearly proving that we can solve industry-relevant optimisation problems that were too hard for quantum computers before. We're now seeing a clear path to a quantum advantage in optimization, and the best part is, it's available to everyone, not just quantum experts."
Global Quantum Intelligence Chief Quantum Analyst David Shaw said: “This is great news for those looking for an early quantum advantage. We're finally working with a number of qubits that could be useful. We expect IBM and others to keep improving the hardware, but it's also important that midstack software is helping more than we thought with specific error suppression and mitigation."
Capgemini's Quantum Lab boss Julian van Velzen said tools like Q-CTRL's on IBM hardware are vital for developers to work together on the noise problem and unlock quantum computing's potential.
“This work shows that using these tools can greatly improve hardware performance and bring a quantum advantage much sooner."