Algorithmic Qubits

Algorithmic Qubits (AQ) is a benchmark IonQ introduced as an alternative to raw physical qubit counts, meant to describe how many qubits a system can actually put to useful work on a real algorithm once connectivity, gate fidelity, and achievable circuit depth are taken into account. The measurement methodology is closely related to IBM's Quantum Volume: both test the largest random circuit a system can run and return a correct result above a fidelity threshold, but IonQ reports the result as a qubit count (#AQ) rather than a volume number, arguing that it is more directly comparable to a system's advertised physical qubit count. Critics of vendor-specific benchmarks note that AQ, like Quantum Volume, is self-reported and uses a methodology the vendor controls, which is why independent third-party benchmarks and application-specific tests remain the more reliable comparison point across different hardware vendors.