Catalog — QPU record · QuEra · Neutral atom
QuEra Aquila
QuEra Aquila is a neutral atom analog processor by QuEra with 256 qubits (2022). Below: every publicly disclosed figure with its source, method, and vendor-claim status — plus what is not disclosed, recorded as findings.
Measurement records7 disclosed · 3 not disclosed · 3 sources
Select a figure to open its record
Qubit count (physical qubit count)256 qubits (maximum number of filled sites)vendor claim
Vendor claim
- Method
- VENDOR_UNSPECIFIED
- Notes
- Whitepaper datasheet also caps total user-defined sites (filled + unfilled) at 256. AWS Braket device page states 'operates up to 256 qubits in analog mode'.
- Citation key
[quera/quera-aquila:physical-qubit-count:2026-08-19]
Coherence (T2* coherence time (Ramsey))5.8 microsecondsvendor claim
Vendor claim
- Source
- Aquila whitepaper datasheet (arXiv:2306.11727) · published 2023-06-20
- Method
- Ramsey protocol on individual non-interacting qubits; includes coherent and incoherent dephasing processes
- Aggregation
- 5.8 us without ensemble averaging; 5.5 us if averaged across the ensemble of atoms in the array (whitepaper text)
- Notes
- Ground-Rydberg qubit coherence, not a hyperfine qubit figure.
- Citation key
[quera/quera-aquila:t2-coherence-time-ramsey:2026-08-19]
Coherence (T2 echo coherence time)11.4 microsecondsvendor claim
Vendor claim
- Source
- Aquila whitepaper datasheet (arXiv:2306.11727) · published 2023-06-20
- Method
- Spin-echo dynamical decoupling protocol; incoherent processes only
- Notes
- Datasheet also lists T2_Rabi = 7.5 us (driven at max Rabi frequency, single qubits) and T2_blockaded-Rabi = 8.9 us (isolated pair of mutually blockaded qubits).
- Citation key
[quera/quera-aquila:t2-echo-coherence-time: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
- quera.com/aquila product page; full Aquila whitepaper datasheet (arXiv:2306.11727, section 1.5) fetched and text-searched — no T1 value is listed; AWS Braket QuEra device page (qualitative 'long lifetimes, supporting tens of qubit flips' only)
Two-qubit gate fidelity (two-qubit gate fidelity)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
- quera.com/aquila; Aquila whitepaper (arXiv:2306.11727); AWS Braket QuEra page
- Notes
- Not applicable as specified: Aquila runs in analog Hamiltonian-simulation mode with no digital gate operations, so no 2Q gate fidelity (or gate duration) is published. See separate blockaded-Rabi proxy fact.
Blockaded-Rabi oscillation quality (weak proxy for Levine-Pichler gate fidelity) (blockaded-Rabi oscillation quality (weak proxy for Levine-Pichler gate fidelity))97.9% probability for 1 atom; 96.0% for 2 mutually blockaded atoms percent (oscillation probability)vendor claim
Vendor claim
- Source
- Aquila whitepaper, Example 2 (arXiv:2306.11727) · published 2023-06-20
- Method
- Rabi oscillations between |gg> and blockaded W-state; whitepaper explicitly calls this 'a weak proxy for the fidelity of the LP gate' (upper bound)
- Notes
- Recorded only because the vendor publishes it as a gate-fidelity proxy; not a gate benchmark.
- Citation key
[quera/quera-aquila:blockaded-rabi-oscillati:2026-08-19]
State preparation and detection errors (state preparation and detection errors)site-fill failure 0.007 (typical); atom detection false-negative 0.01; false-positive 0.01; ground-state mis-detected as Rydberg 0.01; Rydberg mis-detected as ground 0.08 probabilityvendor claim
Vendor claim
- Source
- Aquila whitepaper datasheet (arXiv:2306.11727) · published 2023-06-20
- Method
- Whitepaper datasheet of state preparation, measurement, and Hamiltonian errors; fill probability labeled 'typical' and pattern/site-position dependent
- Aggregation
- typical (as labeled for epsilon_fill); others unlabeled
- Notes
- Readout is via atom presence/absence fluorescence imaging; the asymmetric 0.08 Rydberg-detection error is the dominant readout error.
- Citation key
[quera/quera-aquila:state-preparation-and-de:2026-08-19]
Analog Hamiltonian control ranges (analog Hamiltonian control ranges)max Rydberg Rabi frequency 15.8 rad/us (= 2.5 x 2pi MHz); global detuning |Delta| <= 125 rad/us (approx. +/-20 x 2pi MHz); Rabi slew rate <= 250 rad/us^2; max user-defined evolution 4 us (standard access) as statedvendor claim
Vendor claim
- Source
- Aquila whitepaper datasheet (arXiv:2306.11727); ranges also on quera.com/aquila (Omega 0-2.5 x 2pi MHz, Delta -20 to 20 x 2pi MHz) · published 2023-06-20
- Method
- VENDOR_UNSPECIFIED
- Notes
- Whitepaper notes extended capabilities (e.g., 10 us evolution, 115 um geometries) under 'premium access'.
- Citation key
[quera/quera-aquila:analog-hamiltonian-contr:2026-08-19]
Pricing model (cloud access and pricing model)Amazon Braket (available 'over 100 hours a week' per vendor page); pricing model exists: per-task fee plus per-shot fee on Braketvendor claim
Vendor claim
- Method
- VENDOR_UNSPECIFIED
- Notes
- Launched on Amazon Braket November 2022 (AWS Quantum Technologies Blog: 'Amazon Braket launches Aquila, the first neutral atom quantum processor from QuEra Computing'). QuEra also offers a paid 'Premium Access' tier with direct support.
- Citation key
[quera/quera-aquila:cloud-access-and-pricing:2026-08-19]
One-qubit gate fidelity (one-qubit gate fidelity)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
- quera.com/aquila; Aquila whitepaper datasheet; AWS Braket QuEra page
- Notes
- Analog-mode device; no discrete 1Q gate set or gate fidelity published.
QPU137 is independent and is not affiliated with, endorsed by, or sponsored by QuEra. 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.