BESSER-Q: A Low-Code Path to Quantum Software Development

Quantum technologies are rapidly moving from the lab to national strategies, with multiple governments across Europe are positioning themselves at the forefront of the quantum era. Luxembourg also has an ambition to become “quantum ready” by 2030 and BESSER-Q is here to support that goal.

What is BESSER-Q?

BESSER-Q is an addition to the open-source BESSER platform with visual quantum-circuit modeling and executable code generation in Qiskit targeting three platforms (AER simulation, Noise Mock simulation and IBM hardware). At the backend, everything is stored as instance of a dedicated quantum metamodel. This opens the door for bringing model-based techniques to quantum software development.

BESSER-Q Metamodel

At the heart of BESSER-Q is a dedicated quantum metamodel that bridges quantum computing concepts with broader modeling abstractions of BESSER. By formally linking quantum constructs such as QuantumCircuit to the existing modeling concept of the framework, this metamodel enables quantum components to be directly integrated into hybrid applications generated by BESSER.

The Figure below shows a partial view of BESSER-Q metamodel.

 

The metamodel captures two complementary dimensions of a quantum circuit:

• Structural elements: quantum registers, qubits, classical registers and classical bits.
• Behavioural elements: measurements, single-qubit gates, multi-qubit gates, parameterised gates and user-defined operations.

What BESSER-Q brings to the world of quantum software development?

BESSER-Q provides an end-to-end, model-driven workflow for designing, validating and executing quantum circuits. Its main capabilities include:

• Hardware-agnostic quantum metamodel. The B-UML quantum metamodel represents quantum circuit architectures at an abstract level, independently of a specific simulator or quantum-hardware provider. This lets users focus on the circuit design before selecting an execution target.
• A rich quantum gate catalog. BESSER-Q supports the most common quantum computing building blocks, including single-qubit gates, multi-qubit gates, parameterised gates, measurements, barriers, resets and user-defined operations.
• One model, three execution targets. From the same B-UML circuit model, BESSER-Q can generate Qiskit code for local AER simulation, noise-aware testing with a mock backend, or execution on real IBM Quantum hardware.
• A browser-based visual circuit editor. Users can design circuits graphically in the BESSER Web Modeling Editor by arranging qubits, registers and gates on a canvas. No local installation is required.
• Real-time gate-placement validation. The editor validates circuit construction while users model. It can prevent invalid placements, such as using a multi-qubit operation with too few qubits or applying an operation after a measurement.
• Complete executable-code generation. BESSER-Q generates more than the circuit definition. Generated Qiskit code includes the setup needed for backend selection, transpilation, job submission, result retrieval and classical post-processing.

BESSER Quantum Circuit Editor: First look

The BESSER Quantum editor offers a visual, browser-based workspace for generating quantum code by writing a single line. As shown in the figure above, the center is the circuit canvas, where user can define quantum circuit using the gates shown in the quantum tool box on the left. From the top menu, users can perform action directly on the circuit in editor. The Quality Check feature validates the custom gate using the mathematical properties of the gates and helps identify issues before code is generated. Once the circuit is ready, the Generate menu produces executable Qiskit code for the selected target, whether that is a local simulator, a noise-aware mock backend or IBM Quantum hardware.

What’s Next for Quantum Software Development with BESSER-Q?

One platform for different levels of experts

Today, BESSER-Q operates at circuit level: developers drag and drop gates onto qubit lines in the BESSER Web Modeling Editor. This gives experienced quantum developers direct control over the circuit while avoiding the friction of local tool installation. But the vision goes further. The BESSER-Q roadmap proposes several levels of abstraction so that different communities can participate in quantum software development:

• High level: users express an intended quantum task in plain language, with BESSER generating an inspectable circuit model which is then translated into executable code.
• Medium level: users compose applications from reusable quantum building blocks and patterns.
• Low level: quantum specialists retain full gate-level control over circuits.

More technologies and hardware targets

BESSER-Q currently generates Qiskit code for local AER simulation, noise aware mock simulation and IBM quantum hardware. We plan to broaden this support in future releases by targeting additional quantum software technologies and execution platforms with a more focused approach on Meluxina-Q.

Ultimately, with the addition of BESSER-Q, users can generate hybrid applications with BESSER with traditional, artificial intelligence and quantum components. In such applications, the traditional components can orchestrate workflows, AI components can process or analyse data and quantum component can take care of specialized computational tasks.