The metric for quantum computing success is shifting from qubit count to computational throughput, the number of quantum operations a system can reliably execute each second. Diraq asserts customers ultimately care about results, not qubits themselves, and the cost of those answers is determined by the price of each logical quantum operation.
This price can be calculated for any quantum algorithm, framing quantum computing cost in a way analogous to the “tokens” used in artificial intelligence. “You can calculate the cost of a single logical quantum operation and use it to price any quantum algorithm,” the company reports, suggesting silicon spin qubits offer a favorable cost per operation.
Diraq’s qLDPC Code Enables Problem Pricing & Simulation
Diraq’s qLDPC code allows for preliminary cost calculations on simulations of high-temperature superconductivity and correlated electrons, completing them in under an hour at a cost ranging from a few dollars to a few thousand. This shift in focus allows Diraq to frame the cost of quantum computation in terms of the price of solving specific problems, a strategy similar to the token-based pricing used in artificial intelligence, the company says.
The ability to price quantum algorithms relies on determining the number of reliable logical quantum operations required for a given task, independent of the underlying hardware. Diraq has applied this approach to several computationally intensive problems, revealing the potential for cost-effective solutions previously inaccessible to classical computers. Simulating the quantum chemistry processes needed to improve solar cells, for example, can be completed in hours at a cost of a few thousand dollars using a Diraq system.
Even the notoriously difficult FeMoCo cluster simulation, critical for global fertilizer production, is projected to take weeks and cost around two million dollars on Diraq hardware, a timeframe and price point that could unlock significant advancements in agricultural efficiency. Diraq states that throughput is determined by the number of usable logical qubits and the speed at which logical quantum operations can be performed.
Diraq has run preliminary calculations on the rate of logical operations it expects to achieve with its utility-scale systems using a quantum low-density parity-check (qLDPC) error-correcting code, like the one reported earlier this year by its partner Iceberg Quantum. Once these calculations are complete, price tags can be put on problems instead of on machines. While trapped ion and neutral atom systems excel in certain areas, their slower processing speeds limit overall throughput.
Superconducting and photonic qubits, though fast, face challenges related to scaling and the increasing cost of supporting infrastructure. The qLDPC code is central to Diraq’s approach, enabling the translation of algorithmic demands into quantifiable costs, according to the company. This contrasts with the current industry practice of focusing on qubit count, which provides limited insight into the actual cost of obtaining a solution. The company emphasizes that the number of logical operations required for a problem remains constant regardless of the quantum machine used, with algorithmic improvements being the primary driver of cost reduction.
The feasibility of running an algorithm hinges on meeting a minimum threshold of logical qubits, while commercial viability depends on completing the computation within an acceptable timeframe. Once these parameters are established, the maximum price a customer is willing to pay becomes clear.
This framework allows Diraq to move beyond simply building quantum computers to offering quantum computing as a service, with pricing directly tied to the complexity of the problem and the efficiency of the underlying hardware. The company’s calculations suggest that silicon spin qubits, due to their favorable cost and performance characteristics, are well-positioned to deliver competitive pricing for a wide range of applications, the company says.
Diraq’s methodology highlights a critical distinction between building a quantum computer and using one for practical purposes. The company’s analysis reveals that throughput is not solely dependent on increasing qubit count. While adding more qubits can enable parallel processing, it must be accompanied by a corresponding reduction in per-qubit cost to maintain a competitive price point, even with a large number of qubits, the machine must still be able to complete commercially valuable problems within a reasonable timeframe.
This approach to performance evaluation emphasizes the importance of optimizing both hardware and software to deliver tangible value to customers. Many simulations of new materials or the unknown physical behaviors of solids cannot be performed by classical computers.
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
