Trapped Ion
The highest gate fidelity available in any commercial quantum computer. 56 qubits, ~99.9% two-qubit fidelity, mid-circuit measurement, and the tket compiler built in.
About Quantinuum
Quantinuum was formed in 2021 through the merger of two organizations with deep roots in trapped-ion quantum computing: Honeywell Quantum Solutions, the hardware division of Honeywell, and Cambridge Quantum Computing, a UK-based software company founded by Ilyas Khan. The combined entity brought together hardware expertise and a leading quantum software stack under one roof.
The company uses Ytterbium ions in a cryogenic linear trap with a racetrack architecture for the H2 generation. Ions are shuttled between different zones of the trap: a storage zone, interaction zones where gates are performed, and measurement zones where mid-circuit measurements can be taken without disturbing unmeasured qubits. This architecture enables feed-forward operations where the result of measuring one qubit determines which gates are applied to others in the same circuit.
Quantinuum built tket, one of the most widely used quantum compilers in the industry. pytket provides a unified interface for targeting Quantinuum hardware as well as IBM, IonQ, Google, Rigetti, and simulator backends. tket's optimisation passes reduce gate counts significantly, which is especially valuable on hardware where each gate has a non-trivial fidelity cost.
Hardware Specifications
| Specification | Value |
|---|---|
| Current flagship system | H2-1 |
| Qubit count (H2-1) | 56 qubits |
| Two-qubit gate fidelity (H2-1) | ~99.9% (highest published commercially) |
| Qubit technology | Ytterbium trapped ions (Ba for mid-circuit measurement) |
| Connectivity | All-to-all via shuttling |
| Coherence time | Seconds to minutes |
| Mid-circuit measurement | Yes (feed-forward supported) |
| Just-in-time compilation | Yes (QJIT via tket) |
| Cloud access | Azure Quantum, Quantinuum direct |
| Supported SDKs | tket (pytket), Qiskit, Q#, PennyLane, InQuanto |
Quantinuum's current flagship. 56 qubits arranged in a racetrack ion trap. Holds the record for highest published two-qubit gate fidelity of any commercial quantum computer. Supports mid-circuit measurement and real-time feed-forward, enabling quantum error correction experiments and adaptive circuits.
Earlier H1 systems with 20 qubits and >99.8% two-qubit fidelity. Available for development and smaller workloads. The H1 introduced mid-circuit measurement to commercial cloud access.
Software emulator with a calibrated noise model matching the H-Series hardware. Free tier available for development. Scales to 56 qubits. Use this to develop and validate circuits before running on QPU.
Use Cases
Mid-circuit measurement and real-time feed-forward are required for fault-tolerant QEC protocols like surface codes. H2-1 supports both natively.
InQuanto, Quantinuum's proprietary quantum chemistry toolkit, provides algorithms like TKET-optimised VQE for electronic structure. Runs natively on H-Series.
Feed-forward control allows circuit branches to depend on mid-circuit measurement results, enabling teleportation, magic state distillation, and active error correction.
99.9% gate fidelity and long coherence allow circuits far beyond what lower-fidelity hardware can support, making H-Series the right choice when circuit depth matters.
Quantinuum has published results applying H-Series hardware to pharmaceutical-grade electronic structure problems via InQuanto. See the drug discovery case study.
Researchers publishing fidelity records or quantum volume results frequently use H-Series as the reference platform due to its consistent performance.
Getting Started
The H-Series Emulator is available at no cost through Quantinuum's developer portal. It uses a calibrated noise model from the real hardware, so circuits that pass on the emulator translate reliably to QPU runs.
Quantinuum Developer Portal →Quantinuum built tket (pytket is the Python package), the leading framework-agnostic quantum compiler. The Quantinuum backend extension lets you submit circuits directly.
pip install pytket pytket-quantinuumUse your Quantinuum credentials to authenticate. The QuantinuumBackend class handles job submission, queuing, and result retrieval.
from pytket.extensions.quantinuum import QuantinuumBackend
from pytket import Circuit
backend = QuantinuumBackend(device_name="H2-1E") # E = Emulator
backend.login()
circ = Circuit(2, 2)
circ.H(0).CX(0, 1).measure_all()
compiled = backend.get_compiled_circuit(circ, optimisation_level=2)
handle = backend.process_circuit(compiled, n_shots=100)
result = backend.get_result(handle)
print(result.get_counts())Azure Quantum provides an alternative entry point with credits for new users. Submit Quantinuum circuits using Q#, Qiskit, or pytket through the Azure portal or the azure-quantum Python SDK.
pip install azure-quantum pytket-quantinuumInQuanto is Quantinuum's quantum chemistry software toolkit. It provides pre-built algorithms for electronic structure problems, integrating with pytket and the H-Series backend.
pip install inquantoPricing
Quantinuum uses Hardware Quantum Credits (HQC) as its billing unit. Each quantum gate type has an HQC cost. One- and two-qubit gates have different costs, and the total for a circuit is computed before submission. The emulator is free.
Credits (HQC)
Quantinuum uses Hardware Quantum Credits (HQC). Each gate type consumes a defined number of HQC. Free: H-Series Emulator (H1-1E and H2-1E) available at no cost. QPU credits require a contract or research application.
View pricing details →HQC + Azure Credits
New Azure Quantum accounts receive $500 USD in credits usable across providers including Quantinuum. After credits, HQC are purchased. Enterprise pricing available for sustained workloads.
View pricing details →Learning Resources