NISQ, a term coined by Caltech physicist John Preskill in 2018, describes the current generation of quantum hardware. These devices have enough qubits to be beyond the simulation capability of classical computers (intermediate-scale), but their qubits are too error-prone for fault-tolerant quantum error correction (noisy). The NISQ era poses a fundamental question: can these imperfect machines do anything useful before fully error-corrected quantum computers arrive?
The NISQ landscape encompasses most existing quantum hardware: IBM's Eagle and Heron processors, Google's Sycamore, IonQ's and Quantinuum's trapped-ion systems, and others. Researchers have developed algorithms specifically designed for NISQ constraints, including variational quantum eigensolvers (VQE) for chemistry simulation and the quantum approximate optimization algorithm (QAOA) for combinatorial problems. These hybrid quantum-classical algorithms attempt to extract value from noisy qubits by using classical optimization to compensate for quantum errors.
Results from NISQ algorithms have been mixed. While they can solve small instances of chemistry and optimization problems, classical computers can often match their performance on the same problem sizes. The consensus is shifting toward viewing NISQ as a transitional period rather than a destination — valuable for learning, benchmarking, and building the quantum software ecosystem, but unlikely to deliver transformative commercial advantages. The industry's focus is increasingly on the bridge to fault-tolerant computing, with NISQ devices serving as development platforms. For deeper coverage, see DeepTechIntel's quantum computing section.