Catalog — QPU record · Google · Superconducting

Sycamore

Name: Sycamore specifications
Creator: QPU137

Sycamore is a superconducting gate-based processor by Google with a disclosed qubit count detailed below (2019). Below: every publicly disclosed figure with its source, method, and vendor-claim status — plus what is not disclosed, recorded as findings.

Data as of 2026-08-19Compile a circuit for this class of device Compare

Measurement records7 disclosed · 4 not disclosed · 2 sources

Select a figure to open its record

Qubit count (physical_qubit_count)54 fabricated; 53 functional qubitsvendor claim

Vendor claim

Source: Quantum supremacy using a programmable superconducting processor (Nature 574, 505-510) · published 2019-10-23
Method: VENDOR_UNSPECIFIED
Notes: Peer-reviewed but Google-authored. 'One qubit did not function properly, so the device uses 53 qubits and 86 couplers.'
Citation key: [google/sycamore:physical-qubit-count:2026-08-19]

Single-qubit gate quality (one_qubit_gate_error)0.15% isolated; 0.16% simultaneous % (Pauli error e1)vendor claim

Vendor claim

Source: Quantum supremacy using a programmable superconducting processor (Nature 574) · published 2019-10-23
Method: cross-entropy benchmarking (XEB) at single-qubit level; 'similar to the more typical technique of randomized benchmarking'
Aggregation: mean ('Average (mean) values are shown', Fig. 2)
Notes: Values read directly from Fig. 2a table in the paper PDF.
Citation key: [google/sycamore:one-qubit-gate-error:2026-08-19]

Two-qubit gate quality (two_qubit_gate_error)0.36% isolated; 0.62% simultaneous % (Pauli error e2)vendor claim

Vendor claim

Source: Quantum supremacy using a programmable superconducting processor (Nature 574) · published 2019-10-23
Method: XEB on two-qubit circuits; per-gate e2 isolated by subtracting single-qubit errors e1 from per-cycle error
Aggregation: mean, over all pairs; simultaneous = full-array operation
Notes: Two-qubit per-cycle error e2c: 0.65% isolated / 0.93% simultaneous (Fig. 2a). Simultaneous 0.62% corresponds to ~99.38% mean fidelity for the iSWAP-like gate.
Citation key: [google/sycamore:two-qubit-gate-error:2026-08-19]

Readout (readout_error)3.1% isolated; 3.8% simultaneous %vendor claim

Vendor claim

Source: Quantum supremacy using a programmable superconducting processor (Nature 574) · published 2019-10-23
Method: standard dispersive measurement; simultaneous = all qubits prepared randomly in 0/1 and measured together
Aggregation: mean; 'Measurement errors averaged over the 0 and 1 states'
Citation key: [google/sycamore:readout-error:2026-08-19]

Gate duration (gate_durations)single-qubit 25 ns; two-qubit iSWAP-like 12 ns nsvendor claim

Vendor claim

Source: Quantum supremacy using a programmable superconducting processor (Nature 574) · published 2019-10-23
Method: '25-ns microwave pulses' for 1q; 2q gate via 'turning on a 20-MHz coupling for 12 ns'
Notes: Cirq device docs additionally list SYC gate 12 ns and √iSWAP 32 ns for Sycamore-class devices (https://quantumai.google/cirq/google/devices).
Citation key: [google/sycamore:gate-durations:2026-08-19]

Coherence (T1)not publicly disclosed

Not publicly disclosed — checked 2026-08-19

Finding: No public disclosure found. This is a recorded, dated finding — we do not substitute an estimate or borrow a figure from a sibling device.
Sources queried: Nature 574 main-text PDF (all 6 pages), quantumai.google/cirq/google/devices, Willow spec sheet (covers Willow only)
Notes: Not stated in the Nature 574 main text (pages 505-510, all read this session); coherence data lives in the paper's Supplementary Information, which was not retrieved this session. No vendor spec sheet exists for Sycamore.

Coherence (T2)not publicly disclosed

Not publicly disclosed — checked 2026-08-19

Finding: No public disclosure found. This is a recorded, dated finding — we do not substitute an estimate or borrow a figure from a sibling device.
Sources queried: Nature 574 main-text PDF (all 6 pages), quantumai.google/cirq/google/devices
Notes: Same as T1: not in main text; supplementary not retrieved.

Random circuit sampling benchmark (random_circuit_sampling_benchmark)F_XEB = (2.24 ± 0.21) x 10^-3 at 53 qubits, 20 cycles (elided circuits, Ns = 30M samples); ~200 s to sample one circuit instance one million times vs estimated 10,000 years on a state-of-the-art classical supercomputervendor claim

Vendor claim

Source: Quantum supremacy using a programmable superconducting processor (Nature 574) · published 2019-10-23
Method: linear cross-entropy benchmarking (XEB); classical estimate extrapolated from Summit supercomputer / Google cloud runs
Notes: Peer-reviewed vendor-authored result. The 10,000-year classical estimate was later disputed in the literature (not evaluated here). Largest verified circuits: 53 qubits, 1,113 single-qubit gates, 430 two-qubit gates, predicted total fidelity 0.2%.
Citation key: [google/sycamore:random-circuit-sampling-:2026-08-19]

Cloud access (cloud_access)No public cloud access; Google Quantum Computing Service is approved-list only ('No public access to the service is available at this time').vendor claim

Vendor claim

Source: Quantum Computing Service concepts | Cirq | Google Quantum AI
Method: VENDOR_UNSPECIFIED
Notes: Cirq device docs still describe Sycamore/Sycamore23 layouts alongside Willow105.
Citation key: [google/sycamore:cloud-access:2026-08-19]

Retirement status (retirement_status)not publicly disclosed

Not publicly disclosed — checked 2026-08-19

Finding: No public disclosure found. This is a recorded, dated finding — we do not substitute an estimate or borrow a figure from a sibling device.
Sources queried: quantumai.google (Cirq device/concepts docs), Google Willow blog, Willow early-access guidance PDF, web search for Willow/Sycamore spec pages
Notes: Google has not published a formal retirement notice found this session; Willow is presented as the successor ('~5x improvement over our previous generation of chips', Google blog Dec 9 2024), and the March 2026 early-access guidance is Willow-only, suggesting Sycamore is no longer the offered research device.

Pricing model (pricing_model)not publicly disclosed

Not publicly disclosed — checked 2026-08-19

Finding: No public disclosure found. This is a recorded, dated finding — we do not substitute an estimate or borrow a figure from a sibling device.
Sources queried: quantumai.google/cirq/google/concepts, quantumai.google device docs
Notes: Research-collaboration access only; no public pricing has ever been published.

Topology

square-lattice — '54 transmon qubits, where each qubit is tunably coupled to four nearest neighbours, in a rectangular lattice.' Adjustable couplers; 'One qubit did not function properly, so the device uses 53 qubits and 86 couplers.'

SOURCE: Quantum supremacy using a programmable superconducting processor (Arute et al., Nature 574, 505-510)

Native gate set

1q √X, √Y, √W (25 ns microwave pulses; W=(X+Y)/√2) · PhasedXZ per Cirq device docs
2q iSWAP-like entangling gate, 12 ns ('typical two-qubit gate is a full iSWAP with 1/6th of a full CZ'); exposed in Cirq as SYC = FSimGate(π/2, π/6) · √iSWAP (32 ns, per Cirq device docs)

SOURCE: Quantum supremacy using a programmable superconducting processor (Nature 574); gate-set naming also at https://quantumai.google/cirq/google/devices

Corrections

Found an error in this data? Report it — every report is logged publicly, whether or not it turns out to be right.

QPU137 is independent and is not affiliated with, endorsed by, or sponsored by Google. Figures above are public disclosures recorded with their sources on 2026-08-19; vendor claims are labeled as such and may not be directly comparable across measurement methods — see methodology.