Quantum Circuits are structured sequences or networks of quantum operations applied to Quantum Bits to implement a quantum computation or protocol. A circuit normally specifies the preparation of quantum states, the application of Quantum Gates, interactions between qubits, optional intermediate measurements and final measurement of selected qubits. The circuit model provides a hardware-independent language for describing quantum algorithms while allowing the abstract operations eventually to be compiled into instructions supported by a particular quantum processor.
Quantum circuits are analogous in some respects to classical logic circuits, but they differ because quantum operations are generally reversible and manipulate complex probability amplitudes rather than deterministic Boolean values. A circuit's output is generally probabilistic, meaning that repeated executions can produce a distribution of classical results. The objective is usually to design that distribution so that useful outcomes occur with sufficiently high probability to solve the computational task.
A quantum circuit is commonly represented diagrammatically with horizontal wires corresponding to qubits and symbols on those wires representing operations. Time generally progresses from left to right, although conventions can vary. Single-qubit gates act on individual wires, while multi-qubit gates connect multiple wires to indicate interactions. Measurement operations convert quantum information into classical outcomes and are frequently represented separately from reversible quantum operations.
Initialisation establishes the starting state, often the computational basis state zero for every qubit, although more sophisticated state-preparation procedures can create other states. Classical control may be incorporated into the circuit model, especially when measurement outcomes determine subsequent operations. This makes the circuit model capable of representing both purely unitary computations and protocols involving intermediate measurement, feedback and classical processing.
Circuit depth and gate count are fundamental resource measures. Gate count measures the total number of operations, while depth measures the number of sequential layers required when independent operations can be performed simultaneously. A circuit with many gates can sometimes have relatively low depth if operations are highly parallelisable, whereas a circuit with fewer gates may have greater depth if dependencies force them to occur sequentially.
In noisy hardware, depth is particularly important because qubits must maintain useful coherence throughout the computation. Two-qubit operations frequently contribute disproportionately to error budgets, so circuits with fewer entangling gates may be easier to execute reliably even when their total operation count is not minimal. Other relevant resources include ancilla qubits, classical processing, measurement operations, qubit connectivity and the number of repetitions required to estimate output probabilities.
Quantum circuits can be designed at several abstraction levels. An algorithmic circuit describes high-level mathematical operations, while a logical circuit assumes ideal logical qubits and fault-tolerant operations. A hardware-level circuit uses the native gates, qubit connectivity and constraints of a particular processor. Compilation transforms higher-level circuits into lower-level forms through decomposition and mapping. For example, an abstract controlled operation may need to be represented using several native single-qubit and two-qubit gates. If two logical qubits are not physically adjacent on a hardware topology, additional routing operations may be required. The final circuit can therefore differ substantially from the original algorithmic representation, and the resulting overhead can affect whether a theoretical algorithm is practically executable.
Quantum circuits can implement many distinct computational patterns. State-preparation circuits construct particular quantum states, arithmetic circuits perform operations such as addition or modular arithmetic, and Fourier-transform circuits implement transformations central to several algorithms. Variational circuits contain parameterised gates whose values are adjusted by a classical optimiser and are used in hybrid quantum-classical approaches. Error-correction circuits repeatedly interact data qubits with ancillary qubits to extract syndrome information. Quantum simulation circuits reproduce the dynamics of selected physical systems, while cryptographic circuits can implement mathematical transformations used within specialised protocols. The same circuit formalism can therefore describe algorithms, simulations, communication procedures, error-correction routines and experimental characterisation processes.
Measurement introduces an important distinction between quantum circuit execution and ordinary deterministic program execution. A circuit can produce a probability distribution rather than a single fixed output, so a useful result may require many repeated executions, often called shots, to estimate the distribution accurately. Some algorithms are designed so that the desired answer has high probability after interference has been engineered appropriately, while others estimate expectation values or statistical quantities from repeated measurements.
Measurement basis also matters: applying appropriate basis-change gates before measurement can reveal information about different observables. Mid-circuit measurement can be used for conditional operations, reset, feed-forward and error correction. These features make quantum circuits closely connected to the broader theory of quantum measurement and classical post-processing.
Circuit optimisation is a major practical concern because hardware imperfections make unnecessary operations costly. Compiler techniques can cancel adjacent inverse operations, combine rotations, commute compatible gates, reduce circuit depth, optimise qubit routing and select alternative decompositions. Hardware-aware compilation may account for individual qubit error rates and connectivity, choosing mappings that reduce expected execution error. Dynamic circuits can incorporate measurement results during execution, allowing later operations to depend on information acquired earlier.
Fault-tolerant circuits introduce another layer of complexity because logical operations may require large numbers of physical operations and repeated syndrome measurements. Consequently, an algorithm's abstract circuit complexity can be a poor predictor of practical performance unless the physical implementation and error-correction overhead are also considered.
Quantum Circuits provide the formal bridge between Quantum Algorithms and physical quantum processors. They make quantum computation explicit as a sequence of controllable operations and provide a representation that can be analysed mathematically, optimised by software and translated into hardware instructions. Their effectiveness depends on circuit structure, depth, gate fidelity, connectivity, measurement strategy and error-management requirements.
A circuit that is elegant at the mathematical level may become impractical when mapped to a noisy device, while hardware-aware circuit design can exploit native interactions to reduce overhead. Understanding Quantum Circuits therefore requires both an abstract view based on linear algebra and quantum information and an engineering view based on compilation, control, noise, timing and physical device constraints.
