SoftBank, Quantinuum team on hybrid quantum computing 

Alongside its project Stargate activities, Softbank is working with UK/US firm Quantinuum on hybrid quantum computing systems in data centres.

Combining CPUs, GPUs and Quantum Processing Units (QPUs) holds the potential to further extend the capabilities of AI, says Softbank. The combination means hybrid systems can execute more advanced and diverse computations.

Both companies will use the Japanese market as a foothold to conduct global market research on hybrid quantum computing in the Asia-Pacific region and other regions, and explore specific business models based on that research. 

• Quantinuum is supplying a quantum computer to the Riken Japanese research centre for such a hybrid system.

Softbank sees this being used for quantum chemistry to look for new optical switch materials for All Optical Networks, while in network analysis, it sees the technology being used for anomaly detection and fraud detection in its communication network.

“SoftBank believes in the potential of quantum computers and has been testing and evaluating various internal issues using quantum computers, and has started to obtain certain results. However, as a telecommunications operator, there are still many challenges remaining regarding how to provide quantum computing services in Japan. Through our collaboration with Quantinuum, which possesses the world’s highest-performance quantum computer hardware, we aim to be the first in the world to identify problems that can only be solved by quantum computers and look forward to significantly accelerating the practical application of quantum computing,” said Ryuji Wakikawa, Head of Research Institute of Advanced Technology, SoftBank.

However Softbank says several technical and business challenges need to be addressed to advance the state-of-the-art so that quantum computers are used to solve practical problems. 

The current hardware performance (number of qubits and operation precision) of quantum computers is inadequate for handling practical problems, and significant enhancements in performance are needed for practical use. 

Quantinuum has developed a quantum computer with 32 fully-connected, high-fidelity qubits and an all-new architecture with a new ion trap design that resembles a racetrack. Quantinuum showcased the System Model H2, developed in Munich, Germany, by demonstrating a 32-qubit GHZ state, a non-classical state with all 32 qubits globally entangled. This is the largest on record and is the reason the company says this is the highest performing quantum computer built so far.

The development of hybrid algorithms that combine traditional methods, as well as advancements in error suppression, mitigation, and correction technologies, are essential to enable practical computations.

The concerns also cover the substantial initial investment and operational costs required for the development and deployment of quantum computers that currently lack a concrete return on investment, which in turn suppresses the drive for companies to develop and adopt the technology. 

That is linked to the business models for generating revenue, including the methods of offering quantum computers and setting usage fees.

“Our partnership with SoftBank represents a pivotal moment in the evolution of quantum computing. By combining our strengths, we are poised to unlock innovative solutions that will not only enhance the capabilities of AI but also tackle challenges that have long been beyond reach. Together, we are laying the groundwork for a future where quantum technologies drive transformative advancements across multiple industries,” said Dr. Rajeeb Hazra, President and CEO of Quantinuum. 

SoftBank and Quantinuum say they are committed to addressing these challenges together to advance the practical application of quantum computers. The deal, with the companies jointly considering methods to reduce investment risks, such as revenue sharing and cost sharing.

SoftBank will provide its own business challenges as use cases and both companies will work on quantum chemistry and network analysis, and construct a timeline showing when these use cases will be realized.