Circuit Layer Operations Per Second, a speed benchmark measuring how many quantum circuit layers a processor can execute per unit time including all classical overhead.

What category does CLOPS belong to?

CLOPS is a Performance Metrics concept in quantum computing.

CLOPS — Quantum Computing Glossary

CLOPS (Circuit Layer Operations Per Second) is a throughput benchmark introduced by IBM that measures the end-to-end speed of executing quantum circuits, including compilation, calibration data retrieval, circuit submission, actual quantum execution, and result retrieval. Unlike gate fidelity or quantum volume, which measure computational quality, CLOPS measures computational speed — how quickly a quantum processor can deliver results.

The benchmark runs parameterized circuits at the processor's quantum volume depth and measures the rate at which complete circuit layers are executed. A CLOPS value of 10,000 means the system can execute 10,000 circuit layers per second across the full software-to-hardware stack. IBM's systems have improved from roughly 1,000 CLOPS with early Qiskit Runtime to over 100,000 CLOPS with optimized execution pipelines.

CLOPS matters because many quantum algorithms — particularly variational algorithms like VQE and QAOA — require executing thousands to millions of circuit variants in an iterative optimization loop. A processor with high quantum volume but low CLOPS would take impractically long to run these algorithms. Speed is also critical for quantum error correction, where syndrome measurements must be decoded in real time (typically within microseconds) to apply corrections before the next round of stabilizer measurements. The classical processing speed for syndrome decoding is an increasingly important bottleneck as processors scale toward fault tolerance.