Surface-17

Raveh Neeman

doi:10.5281/zenodo.22674814

ETH Zurich (Quantum Device Lab) · superconducting circuits · DEMONSTRATED (2022-2025) · 17 physical qubits · access: lab-only

architecture: Transmon lattice with tunable couplers · Open on the map

Register profile — the machine's variant of its architecture

qubit type: flux-tunable transmon (asymmetric SQUID)

gate mechanism: flux-pulsed CZ between directly coupled neighbours (no tunable coupler)

connectivity: diagonal planar square lattice, nearest-neighbour

control: room-temperature AWGs and UHFQA; IQ up/down-conversion; FPGA demodulation

control placement: room-temp

readout: frequency-multiplexed dispersive; readout resonator + Purcell filter per qubit; TWPA; three-state discrimination flags leakage

gate evidence: a published figure (source class not yet tagged)

decoder mode: offline — syndromes decoded after the run

codes: quantum_repetition, surface-17

Technologies by layer — the machine's cell per layer

1 Qubit carrier	Transmon · primary ▣ Fig. 1b-c; Methods A; Table I ✅ 17 flux-tunable transmons with asymmetric SQUIDs: 9 data + 8 auxiliary; mean T1 32.5 µs
2 Encoding	Bare physical qubit (no encoding) · primary ◌ 🔎 bare transmon two-level subspace; the third level is read only to flag leakage
3 Gate mechanism	Tunable-coupler CZ / iSWAP · primary ▣ Methods A; Appendix B (Fig. 5); Appendix C (Fig. 6a) 🔎 CZ by tuning adjacent data/auxiliary pairs into resonance with net-zero flux pulses on individual flux lines; average CZ error 1.5(1.0)%, best 0.6%
4 Connectivity / transport	Static nearest-neighbour lattice · primary ▣ Fig. 1a-b ✅ static diagonal planar square lattice; each auxiliary couples to up to four data qubits
5 Control	Room-temperature electronics + per-qubit coax/flex · primary ▣ Appendix D (Fig. 9) ✅ room-temperature AWGs (2.4 GSa/s) for flux, UHFQA (1.8 GSa/s) for readout, IQ up/down-conversion with FPGA demodulation
6 Readout	Dispersive microwave readout (+TWPA, Purcell) · primary ▣ Fig. 1b,d; Methods B; Appendix F; Table I ✅ frequency-multiplexed dispersive readout through a readout resonator + Purcell filter per qubit and a TWPA; 0.9% mean two-state error
7 Code	Rotated surface code (+ yoked variants) · primary ▣ Fig. 1a; Fig. 3a; Fig. 4c ✅ distance-3 rotated surface code (Surface-17): 1.1 µs cycles, 3.2(1)% logical error per cycle with leakage rejection (Nature 2022); later split by lattice surgery into two d=3 repetition codes (arXiv Jan 2025)
8 Decoder	MWPM / Sparse Blossom (+ correlated matching) · primary ▤ Abstract; Appendix L ✅ minimum-weight perfect matching with weights from syndrome correlations (error-model-free); corrections applied in post-processing
9 Interconnect	— none — nothing in this layer ▣ Fig. 1b; Appendix A ✅ single 17-qubit chip; no inter-module link
10 Manufacturing	Superconducting-qubit lithography (Al/AlOx/Al junctions, Nb wiring, 300 mm) · primary ▤ Appendix A ✅ 150 nm Nb on high-resistivity Si (photolithography + reactive-ion etching), Al Josephson junctions by e-beam lithography, Al-Ti-Al airbridges

evidence: ✅ 8 / 10