D-Wave has opened beta access to the gate-model quantum computing simulator it announced in June. The beta gives select commercial and research organizations early access to experiment with error-aware programming and develop quantum applications for D-Wave’s forthcoming gate-model systems, the company said.

The simulator uses D-Wave’s dual-rail superconducting gate-model architecture, which D-Wave says is designed to make certain errors easier to detect and reduce the hardware needed for quantum error correction as systems scale.

The simulator is available through D-Wave’s Leap cloud service and Ocean SDK. Germany’s Jülich Supercomputing Centre and Florida Atlantic University are among the beta participants, as well as the Spanish banking group BBVA and the Stockholm-based quantum technology company FirstQFM.

D-Wave’s documentation lists simulator configurations for 17 and 21 qubits, with 21 qubits as the current maximum. The software can simulate detectable errors and supports real-time control features that D-Wave says it is still phasing into its physical gate-model systems. Importantly, this is a classical Monte Carlo simulator built on Qiskit’s AerStatevector software, so users are not accessing a physical gate-model quantum processor.

Even without access to physical quantum hardware, the simulator could give developers a significant head start. They can begin building applications, testing how programs respond to detected errors and working with the real-time control features D-Wave plans to carry into its gate-model systems.

Beta participants will use the simulator for different types of research and application development. BBVA said it plans to explore financial applications including portfolio optimization, fraud detection and derivatives valuation. FirstQFM plans to test whether detected errors, mid-circuit error signals and real-time control can improve the quality and reliability of quantum computations.

At FAU, researchers will study how quantum machine learning models behave under more realistic hardware constraints and whether error-aware techniques can make them more reliable. Jülich Supercomputing Centre, which already hosts D-Wave quantum annealing systems as part of its JUNIQ infrastructure, said it will use early access to explore error-aware quantum computing and provide input toward D-Wave’s future gate-model systems.

D-Wave has not given a general availability date or announced standalone pricing for the simulator. It previously said it plans development bundles combining simulator access, future gate-model systems and technical guidance, with bundle pricing available on request.