State of Quantum Hardware

A cross-vendor ranking of quantum devices, built from the public corpus of calibration snapshot s and openly-licensed benchmarks. Switch the ranking metric below. Rows are labelled by source & licence — vendor-reported rows are manufacturer claims, not independently reproduced here. How the corpus works →

Best 2Q fidelity 2Q gate fidelity Median 2Q error Median T1 (µs) Median T2 (µs) Algorithmic qubits (#AQ) Quantum Volume CLOPS EPLG (error/layer) Bell-state effective qubits QFT accuracy QAOA approx. ratio Qubits
Rank Provider Backend QAOA approx. ratio Source / licence Captured
#1 quantinuum h2-2 0.8018 Metriq · CC-BY-4.0 2026-01-23
#2 ibm ibm_boston 0.6734 Metriq · CC-BY-4.0 2026-02-18
#3 ibm ibm_kingston 0.6412 Metriq · CC-BY-4.0 2026-02-17
#4 ibm ibm_marrakesh 0.5731 Metriq · CC-BY-4.0 2026-02-17
#5 ibm ibm_pittsburgh 0.5695 Metriq · CC-BY-4.0 2026-02-19
#6 ibm ibm_torino 0.456 Metriq · CC-BY-4.0 2026-02-19
#7 ibm ibm_fez 0.4209 Metriq · CC-BY-4.0 2026-02-17
#8 iqm iqm_garnet 0.1432 Metriq · CC-BY-4.0 2026-03-02
#9 iqm iqm_emerald 0.0099 Metriq · CC-BY-4.0 2025-11-19
#10 origin wukong_72 -0.035 Metriq · CC-BY-4.0 2026-02-24
Sources & attribution. IBM rows are real device calibration snapshots from Qiskit's fake_provider (Apache-2.0). Metriq rows are community benchmark submissions from metriq.info (CC-BY-4.0). Zenodo rows are openly-licensed raw calibration datasets (CC-BY/CC0), cited per row. Vendor-reported rows are specifications published by the manufacturer and are shown as claims, with the press/source link recorded in each snapshot's provenance. Metrics differ in what they measure and how they're obtained, so compare within a metric and mind the source badge.