← Quantum Technology AtlasTechnologies · Machines · Architectures · OrganisationsРусскийעברית

Sycamore

Google Quantum AI

Sycamore

Google Quantum AI · superconducting circuits · RETIRED (demonstrated Oct 2019) · 53 physical qubits · access: lab-only

Register profile — the machine's variant of its architecture

qubit type: frequency-tunable transmon (asymmetric SQUID, Al junctions)
gate mechanism: 86 tunable couplers; native iSWAP-like fSim, 12 ns; 1Q 25 ns
connectivity: square lattice, degree 4
control: 277 DACs (14-bit, 1 GS/s) at room temperature
control placement: room-temp
readout: frequency-multiplexed dispersive
gate evidence: a published figure (source class not yet tagged)
codes: css_4_1_2, quantum_repetition, rotated_surface, surface
flags: claim later eroded

Technologies by layer — the machine's cell per layer

1 Qubit carrier
Transmon · primary ▣ Fig. 1a-b (54-qubit layout; chip photograph); SI §II Fabrication and layout ✅
frequency-tunable transmon (asymmetric SQUID), Al junctions; 54 fabricated, 53 operational; T1 ≈ 20 µs (2019)
2 Encoding
Bare physical qubit (no encoding) · primary ◌ 🔎
bare transmon two-level subspace; no encoding beyond the physical qubit
3 Gate mechanism
Tunable-coupler CZ / iSWAP · primary ▣ Fig. 1a (couplers, blue); Fig. 3b (1Q/2Q gate control waveforms); SI §II (88 adjustable couplers) ✅
86 tunable couplers; iSWAP-like fSim, ~20 MHz on resonance for 12 ns + CPhase; 1Q 25 ns; 2Q Pauli 0.36% isolated, 0.62% simultaneous
4 Connectivity / transport
Static nearest-neighbour lattice · primary ▣ Fig. 1a (rectangular array, each qubit coupled to four nearest neighbours) ✅
square lattice, degree 4; 53 operational qubits, 86 couplers
5 Control
Room-temperature electronics + per-qubit coax/flex · primary ▣ SI Fig. S2 (control electronics, DAC module); Fig. S3 (cryogenic wiring); Fig. S4 (chassis); SI §III.1 ✅
277 DACs (14-bit, 1 GS/s) at room temperature; ≈5.2 lines per qubit (53 q)
6 Readout
Dispersive microwave readout (+TWPA, Purcell) · primary ▣ SI Fig. S2c (readout chain: AWG, amplification, IF mixing, ADCs); SI §III.2 Readout ✅
frequency-multiplexed dispersive readout
7 Code
Error-detection codes ([[4,2,2]], d=2 surface, iceberg, spacetime) · primary ▤ abstract; Fig. 4 ✅
repetition codes d=3–11 for 50 rounds (exponential suppression); d=2 surface code [[4,1,2]] run as error detection (sweep 7 Oct 2026, source read)
8 Decoder
— none — nothing in this layer ▤ 🔎
no decoder; NISQ device, no QEC loop
9 Interconnect
Multi-die packaging in one module (flip-chip, bump-bonded control die) · primary ▤ §Building a high-fidelity processor (indium bump-bonds between two Si wafers); SI Fig. S1 (packaged chip) ✅
one qubit die + control die, flip-chip with indium bump-bonds between two silicon wafers; no inter-module interconnect
10 Manufacturing
Superconducting-qubit lithography (Al/AlOx/Al junctions, Nb wiring, 300 mm) · primary ▤ §Building a high-fidelity processor (Al metallization + JJs, indium bump-bonds); SI §II Fabrication and layout ✅
Al Josephson junctions; two-wafer flip-chip with indium bump-bonds (packaging lineage Google still uses)

Standard records

2Q error under full-width parallel operation → 6.20e-3 · typical (0.62% simultaneous)
Oct 2019 · source
control lines per qubit → 5.2 lines/qubit · typical (277 DACs / 53 q)
2019 · source
1Q gate time → 25 ns · best
Oct 2019 · source
evidence: ✅ 8 / 10

Pictures

Every picture carries the story of what it shows and its credit line.

File:Google Sycamore Chip 001.png
Still frame from a video showing a two-chip stack under blue light: a small dark die mounted face-down on a larger mirror-like die whose fine lines fan out to its edges, resting on a machined aluminium pedestal. The caption says it comes from a Google video (2019-10-23) showing the Sycamore quantum processor. It illustrates flip-chip assembly of a superconducting processor: two dies are bonded face to face so wiring can run on a separate layer; the radial lines are those signal traces. “Google Sycamore Chip 001” by Google (www.youtube.com/channel/UCK8sQmJBp8GCxrOtXWBpyEA), commons.wikimedia.org/wiki/File:Google_Sycamore_Chip_001.png, CC BY 3.0 (creativecommons.org/licenses/by/3.0/)
File:Sycamore Quantum Chip.jpg
A photograph of Google's Sycamore quantum chip on display at the Deutsches Museum, Munich: a thick brushed-aluminium mount engraved 'Google' and 'Sycamore', holding a light-grey carrier die with radial wiring lines and a smaller dark die mounted face-down on it. Photographed by Coldupnorth (November 2025). Sycamore is a superconducting processor built from transmon qubits; the thing to notice is the two-die flip-chip stack, since individual qubits are too small to see at this resolution. “Sycamore Quantum Chip” by Coldupnorth, commons.wikimedia.org/wiki/File:Sycamore_Quantum_Chip.jpg, CC BY SA 4.0 (creativecommons.org/licenses/by-sa/4.0/)
File:Google Sycamore Chip 002.png
A computer-animation frame, not a photograph: a rectangular carrier board with hundreds of fine wiring traces radiating from its centre and a smaller dark square die stacked on top, under the title 'Sycamore Chip'. It is a screenshot from a Google video (2019) about its Sycamore quantum processor. The render illustrates the flip-chip stacking used in superconducting processors, where a smaller die is bonded face-down onto a larger wiring chip that fans control lines out; real surface detail is idealised away. “Google Sycamore Chip 002” by Google (www.youtube.com/channel/UCK8sQmJBp8GCxrOtXWBpyEA), commons.wikimedia.org/wiki/File:Google_Sycamore_Chip_002.png, CC BY 3.0 (creativecommons.org/licenses/by/3.0/)