Catalog — QPU record · IonQ · Trapped ion
IonQ Forte
IonQ Forte is a trapped ion gate-based processor by IonQ with 36 qubits (2022). Below: every publicly disclosed figure with its source, method, and vendor-claim status — plus what is not disclosed, recorded as findings.
Measurement records12 disclosed · 0 not disclosed · 4 sources
Select a figure to open its record
Qubit count (physical_qubit_count)36 qubitsvendor claim
Vendor claim
- Method
- VENDOR_UNSPECIFIED
- Notes
- Corroborated by Azure Quantum docs (36 qubits, ionq.qpu.forte-1). The peer-reviewed benchmarking paper (Quantum 8, 1516 (2024)) used a 30-qubit single-chain configuration of Forte.
- Citation key
[ionq/ionq-forte:physical-qubit-count:2026-08-19]
Algorithmic qubits (#AQ) (algorithmic_qubits_AQ)35 #AQvendor claim
Vendor claim
- Source
- IonQ Achieves Technical Milestone One Year Ahead of Schedule (#AQ 35 on Forte) · published 2024-01-25
- Method
- Derived from algorithmic benchmarking protocols established by QED-C industry study; most complex circuits passed (qubits x gates) determine the #AQ score
- Notes
- Peer-reviewed (IonQ-authored) paper arXiv:2308.05071 / Quantum 8, 1516 (2024) validated #AQ 29 on the 30-qubit Forte configuration, using error mitigation by symmetrization (25 circuit variants). #AQ is distinct from the 36 physical qubits.
- Citation key
[ionq/ionq-forte:algorithmic-qubits-aq:2026-08-19]
Two-qubit gate quality (two_qubit_gate_fidelity)99.6% (stated as 0.4% 2Q gate error) %vendor claim
Vendor claim
- Method
- Direct Randomized Benchmarking (DRB)
- Aggregation
- not stated on spec page
- Notes
- Consistent with the peer-reviewed paper's median (see separate fact).
- Citation key
[ionq/ionq-forte:two-qubit-gate-fidelity:2026-08-19]
Two-qubit gate quality (two_qubit_gate_error_distribution_peer_reviewed)median 46.4e-4 (0.464%); 10th percentile 34.5e-4; 90th percentile 99.6e-4; best pair 27.8e-4; worst pair 885e-4 error probability (parts per ten thousand as reported)vendor claim
Vendor claim
- Source
- Benchmarking a trapped-ion quantum computer with 30 qubits (Quantum 8, 1516 (2024)) · published v1 2023-08-09; v2 2024-11-02
- Method
- Direct Randomized Benchmarking across all C(30,2)=435 gate pairs of a single-chain 30-qubit Forte, collected over ~6 months
- Aggregation
- median, with 10th/90th percentiles, best and worst pair
- Notes
- IonQ-authored (Chen, Nielsen, ... Gamble) but peer-reviewed in Quantum journal; marked vendor claim because measured by the vendor, not a third party.
- Citation key
[ionq/ionq-forte:two-qubit-gate-error-dis:2026-08-19]
Single-qubit gate quality (one_qubit_gate_fidelity)99.98% (stated as 0.02% 1Q gate error) %vendor claim
Vendor claim
- Method
- Clifford Randomized Benchmarking (vendor page); paper used randomized benchmarking on the 30-qubit chain
- Aggregation
- not stated on spec page; paper reports median 1Q error 2.0e-4 (10th pct 1.8e-4, 90th pct 2.6e-4)
- Notes
- Paper figures from arXiv:2308.05071 (Quantum 8, 1516 (2024)).
- Citation key
[ionq/ionq-forte:one-qubit-gate-fidelity:2026-08-19]
Readout (readout_SPAM_fidelity)99.5% (0.5% SPAM error) %vendor claim
Vendor claim
- Method
- state preparation and measurement error
- Aggregation
- average ('0.5% on each qubit across the entire ion chain' per paper)
- Notes
- Same 0.5% figure in arXiv:2308.05071.
- Citation key
[ionq/ionq-forte:readout-spam-fidelity:2026-08-19]
Coherence (T1)10-100 svendor claim
Vendor claim
- Method
- VENDOR_UNSPECIFIED
- Citation key
[ionq/ionq-forte:t1:2026-08-19]
Coherence (T2)~1 svendor claim
Vendor claim
- Method
- VENDOR_UNSPECIFIED
- Citation key
[ionq/ionq-forte:t2:2026-08-19]
Gate duration (two_qubit_gate_duration)~900 microsecondsvendor claim
Vendor claim
- Source
- Benchmarking a trapped-ion quantum computer with 30 qubits (Quantum 8, 1516 (2024)) · published v2 2024-11-02
- Method
- as reported in peer-reviewed benchmarking paper
- Aggregation
- average (30-qubit chain configuration)
- Notes
- Not published on the vendor spec page; taken from the paper (2Q duration depends on chain length).
- Citation key
[ionq/ionq-forte:two-qubit-gate-duration:2026-08-19]
Single-qubit gate quality (one_qubit_gate_duration)110 microsecondsvendor claim
Vendor claim
- Source
- Benchmarking a trapped-ion quantum computer with 30 qubits (Quantum 8, 1516 (2024)) · published v2 2024-11-02
- Method
- nine pi/2 pulses using SK1 dynamical correction (per paper)
- Citation key
[ionq/ionq-forte:one-qubit-gate-duration:2026-08-19]
Pricing model (pricing_model_exists)Yes - AWS Braket on-demand: $0.30 per task + $0.08 per shot; reservation $7,000.00/hour; minimum 2,500 shots per task when using error mitigation. Azure Quantum: Pay-As-You-Go metered in qubit-gate-shots (QGS), debiasing-based pricing (debiasing on by default) USDverified
Independently corroborated
- Source
- Amazon Braket Pricing
- Method
- VENDOR_UNSPECIFIED
- Notes
- Azure details from learn.microsoft.com/en-us/azure/quantum/provider-ionq.
- Citation key
[ionq/ionq-forte:pricing-model-exists:2026-08-19]
Release year (release_year)2022 ('Available since 2022' per vendor page; 'Introducing IonQ Forte' announcement May 2022)vendor claim
Vendor claim
- Method
- VENDOR_UNSPECIFIED
- Notes
- Announcement: businesswire.com/news/home/20220517005569/en (May 17, 2022, seen as search result title; not fetched).
- Citation key
[ionq/ionq-forte:release-year:2026-08-19]
QPU137 is independent and is not affiliated with, endorsed by, or sponsored by IonQ. 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.