System overview transmon backend

The transmons backend comprises of two transmon-based quantum processors: Tuna-5 and Starmon-7. Tuna-5: The Tuna-5 system features a quantum processor consisting of 5 transmons in a starfish configuration, with 4 dedicated tunable couplers connecting the central transmon Q0 to each of the 4 corner ones. The use of tunable couplers in the quantum processors lends the name "Tuna" to this quantum computing system. Starmon-7: At the core of our Starmon-7 back-end lies a superconducting quantum processor based on circuit quantum electrodynamics consisting of seven superconducting transmon qubits. The high connectivity of the transmon qubits in this chip earns them the nickname Starmon. Qubit operations are performed with room-temperature analog and digital control systems programmed using a high-level language and compiler. Extensibility to larger qubit systems is a key feature in the design of every layer of this full stack.

• 

  Algorithm

  Quantum Inspire

  Quantum Inspire's user experience comes with a web portal with an intuitive editor for novice users and a software development kit (SDK) to program algorithms, execute these algorithms and examine the results in various ways. Account management, job and result management is available for all registered users.

• 

  Programming language

  cQASM and SDK

  The programming language of Quantum Inspire platform is cQASM, developed at QuTech. Quantum algorithms can be programmed in cQASM using the QI online editor. The SDK provides an interface between the QI Application Programming Interface (API) and Python-based quantum programming platforms (e.g. QisKit and Pennylane ) and performs any translation between these languages.

• 

  OpenSquirrel

  Compiler

  OpenSquirrel is a quantum compiler that chooses a modular, over a configurable, approach to prepare and optimize quantum circuits for heterogeneous target architectures. It has a user-friendly interface and is straightforwardly extensible with custom-made readers, compiler passes, and exporters. As a quantum circuit compiler, it is fully aware of the semantics of each gate and arbitrary quantum gates can be constructed manually. It supports the cQASM quantum programming language, using libQASM as a language parser. It is developed in modern Python and follows best practices. The compiler produces intermediate quantum assembly language, a hardware-agnostic representation of a circuit that can be simulated and run directly on quantum hardware. Compilation starts by decomposing quantum operations into physical instructions. Then, the compiler performs optimizations such as merging of single-qubit rotations. After these steps, the compiler analyzes the dependencies of quantum gates and and schedules instructions respecting the duration of gates. For Tuna-5 system OpenSquirrel generates a quantify schedule to send instructions to the room-temperature control electronics of the Tuna-5 system.

  For the Starmon-7 backend, OpenSquirrel also uses the legacy OpenQL compiler for some backward compatibility issues.

• 

  Classical to quantum

  Control Electronics

  Modularity was a key design feature for the control electronics used in the Tuna-5 and Starmon-7 back-ends.

  For Tuna-5 system all control and readout signals are generated and acquired at room temperature using a Qblox Cluster that contains one QRM-RF module for readout, three QCM-RF modules for single-qubit gates and three QCM modules for baseband flux control of transmons and couplers. 

  For the Starmon-7 system, custom digital and analog control systems were created to shape and time the microwave-frequency and baseband pulses required to perform qubit operations. Room-temperature microwave sources, mixers and arbitrary waveform generators (AWGs) are used to generate and steer the microwave pulses required to perform single-qubit gates and readout. AWGs are also used to generate the baseband signals needed for two-qubit gates. The QuTech Central Controller orchestrates the operation of all these analog instruments.  Signals inputs are sent to the quantum chip via heavily filtered coaxial lines and signal outputs are conditioned using cryogenic superconducting and semiconducting amplifiers.

• 

  Tuna-5

  Quantum Processor

  Tuna-5 consists of 5 transmons in a starfish configuration, with 4 dedicated tunable couplers connecting the central transmon Q0 to each of the 4 corner ones. Every transmon has a dedicated microwave-drive line and fluxcontrol line, as well as a dedicated pair of readout Purcellfilter resonators for independent readout via one common feedline. Tunable couplers have dedicated flux-control lines. The otherwise planar device, fabricated from a Ta base layer sputtered on a Si substrate, has Al crossovers and airbridges, including ’shoelacing’ airbridges enabling postfabrication frequency trimming of readout resonator and Purcell filters. The device earns its name "Tuna" from the use of flux-tunable qubit couplers. Tuna-5 processor was fabricated by QuantWare using the Hamiltonian design provided by the DiCarlo lab at QuTech, TU Delft. 

  Starmon-7 consists of 7 transmon qubits in a circuit quantum electrodynamics chip architecture. Similar devices and the same control architecture have been previously used by the DiCarlo lab in QuTech to realize quantum error detection [Marques2022], quantum neural networks [Moreira2023], and variational algorithms [Sagastizabal2021]. The device connectivity allows native fast two-qubit gates between nearest-neighbor qubits (8 pairs). These gates are realized using dedicated flux-control lines available for each qubit. Additionally, each qubit has a dedicated microwave-control line for single-qubit gating. Finally, each qubit is independently readout using a dedicated dispersively-coupled readout resonator with dedicated Purcell filter. These readout structures connect to two feedlines allowing simultaneous readout by frequency multiplexing.

  [Marques2022] J.F. Marques et al., Logical-qubit operations in an error-detecting surface code, Nature Physics 18, 80 (2022).

  [Moreira2023] M.S. Moreira et al., Realization of a quantum neural network using repeat-until-success circuits in a superconducting quantum processor, NPJ Quantum Information 9, 118 (2023).

  [Sagastizabal2021] R. Sagastizabal et al., Variational preparation of finite-temperature states on a quantum computer, NPJ Quantum Information 7, 130 (2021).

  [Vallés-Sanclemente2025] S. Vallés-Sanclemente et al., Optimizing the frequency positioning of tunable couplers in a circuit QED processor to mitigate spectator effects on quantum operations, (pre-print) https://arxiv.org/pdf/2503.13225 (17 March 2025)