NISQ (Noisy Intermediate-Scale Quantum) was coined by physicist John Preskill in 2018 to describe the current era of quantum hardware. NISQ devices have 50 to ~1000 qubits, but they are "noisy" — gate errors, decoherence, and readout errors mean circuits cannot be run indefinitely. NISQ algorithms are designed to work within these constraints: they use shallow circuits and hybrid classical-quantum approaches to avoid accumulating too much noise. Notable NISQ algorithms include VQE (Variational Quantum Eigensolver) and QAOA (Quantum Approximate Optimization Algorithm). Most publicly accessible quantum computers today — IBM, IonQ, Quantinuum — are NISQ devices. Post-NISQ hardware with full fault tolerance (requiring millions of physical qubits) is still many years away.
Related Terms
Decoherence
HardwareThe loss of quantum properties when a qubit interacts with its environment.
Quantum Error Correction
HardwareTechniques to detect and correct errors in quantum circuits without measuring (and collapsing) the qubits.
VQE
AlgorithmsVariational Quantum Eigensolver — a hybrid quantum-classical algorithm for finding ground state energies.
QAOA
AlgorithmsQuantum Approximate Optimization Algorithm — a hybrid algorithm for combinatorial optimization problems.
Quantum Volume
MetricsIBM's single-number benchmark measuring the overall capability of a quantum computer, accounting for qubits, connectivity, and fidelity.