This list contains quantum processors, also known as quantum processing units (QPUs). Some devices listed below have only been announced at press conferences so far, with no actual demonstrations or scientific publications characterizing the performance.
Quantum processors are difficult to compare due to the different architectures and approaches. Due to this, published physical qubit numbers do not reflect the performance levels of the processor. This is instead achieved through the number of logical qubits or benchmarking metrics such as quantum volume, randomized benchmarking or circuit layer operations per second (CLOPS). [1]
These QPUs are based on the quantum circuit and quantum logic gate-based model of computing.
Manufacturer | Name/codename designation | Architecture | Layout | Fidelity (%) | Qubits (physical) | Release date | Quantum volume |
---|---|---|---|---|---|---|---|
Alpine Quantum Technologies | PINE System [2] | Trapped ion | 24 [3] | June 7, 2021 | 128 [4] | ||
Atom Computing | Phoenix | Neutral atoms in optical lattices | 100 [5] | August 10, 2021 | |||
Atom Computing | N/A | Neutral atoms in optical lattices | 35×35 lattice (with 45 vacancies) | < 99.5 (2 qubits) [6] | 1180 [7] [8] | October 2023 | |
N/A | Superconducting | N/A | 99.5 [9] | 20 | 2017 | ||
N/A | Superconducting | 7×7 lattice | 99.7 [9] | 49 [10] | Q4 2017 (planned) | ||
Bristlecone | Superconducting transmon | 6×12 lattice | 99 (readout) 99.9 (1 qubit) 99.4 (2 qubits) | 72 [11] [12] | March 5, 2018 | ||
Sycamore | Superconducting transmon | 9×6 lattice | N/A | 53 effective (54 total) | 2019 | ||
Willow | Superconducting transmon | 7×7 lattice | 99.965% (1-qubit) 99.67% (2-qubit) Surface code error correction implemented. | 49 effective (105 total) | December 9, 2024 [13] | ||
IBM | IBM Q 5 Tenerife | Superconducting | bow tie | 99.897 (average gate) 98.64 (readout) | 5 | 2016 [9] | |
IBM | IBM Q 5 Yorktown | Superconducting | bow tie | 99.545 (average gate) 94.2 (readout) | 5 | ||
IBM | IBM Q 14 Melbourne | Superconducting | N/A | 99.735 (average gate) 97.13 (readout) | 14 | ||
IBM | IBM Q 16 Rüschlikon | Superconducting | 2×8 lattice | 99.779 (average gate) 94.24 (readout) | 16 [14] | May 17, 2017 (Retired: 26 September 2018) [15] | |
IBM | IBM Q 17 | Superconducting | N/A | N/A | 17 [14] | May 17, 2017 | |
IBM | IBM Q 20 Tokyo | Superconducting | 5×4 lattice | 99.812 (average gate) 93.21 (readout) | 20 [16] | November 10, 2017 | |
IBM | IBM Q 20 Austin | Superconducting | 5×4 lattice | N/A | 20 | (Retired: 4 July 2018) [15] | |
IBM | IBM Q 50 prototype | Superconducting transmon | N/A | N/A | 50 [16] | ||
IBM | IBM Q 53 | Superconducting | N/A | N/A | 53 | October 2019 | |
IBM | IBM Eagle | Superconducting transmon | N/A | N/A | 127 [17] | November 2021 | |
IBM | IBM Osprey [7] [8] | Superconducting | N/A | N/A | 433 [17] | November 2022 | |
IBM | IBM Condor [18] [7] | Superconducting | Honeycomb [19] | N/A | 1121 [17] | December 2023 | |
IBM | IBM Heron [18] [7] | Superconducting | N/A | N/A | 133 | December 2023 | |
IBM | IBM Heron R2 [20] | Superconducting | Heavy hex | 96.5 (2 qubits) | 156 | November 2024 | |
IBM | IBM Armonk [21] | Superconducting | Single Qubit | N/A | 1 | October 16, 2019 | |
IBM | IBM Ourense [21] | Superconducting | T | N/A | 5 | July 3, 2019 | |
IBM | IBM Vigo [21] | Superconducting | T | N/A | 5 | July 3, 2019 | |
IBM | IBM London [21] | Superconducting | T | N/A | 5 | September 13, 2019 | |
IBM | IBM Burlington [21] | Superconducting | T | N/A | 5 | September 13, 2019 | |
IBM | IBM Essex [21] | Superconducting | T | N/A | 5 | September 13, 2019 | |
IBM | IBM Athens [22] | Superconducting | N/A | 5 | 32 [23] | ||
IBM | IBM Belem [22] | Superconducting | Falcon r4T [24] | N/A | 5 | 16 [24] | |
IBM | IBM Bogotá [22] | Superconducting | Falcon r4L [24] | N/A | 5 | 32 [24] | |
IBM | IBM Casablanca [22] | Superconducting | Falcon r4H [24] | N/A | 7 | (Retired – March 2022) | 32 [24] |
IBM | IBM Dublin [22] | Superconducting | N/A | 27 | 64 | ||
IBM | IBM Guadalupe [22] | Superconducting | Falcon r4P [24] | N/A | 16 | 32 [24] | |
IBM | IBM Kolkata | Superconducting | N/A | 27 | 128 | ||
IBM | IBM Lima [22] | Superconducting | Falcon r4T [24] | N/A | 5 | 8 [24] | |
IBM | IBM Manhattan [22] | Superconducting | N/A | 65 | 32 [23] | ||
IBM | IBM Montreal [22] | Superconducting | Falcon r4 [24] | N/A | 27 | 128 [24] | |
IBM | IBM Mumbai [22] | Superconducting | Falcon r5.1 [24] | N/A | 27 | 128 [24] | |
IBM | IBM Paris [22] | Superconducting | N/A | 27 | 32 [23] | ||
IBM | IBM Quito [22] | Superconducting | Falcon r4T [24] | N/A | 5 | 16 [24] | |
IBM | IBM Rome [22] | Superconducting | N/A | 5 | 32 [23] | ||
IBM | IBM Santiago [22] | Superconducting | N/A | 5 | 32 [23] | ||
IBM | IBM Sydney [22] | Superconducting | Falcon r4 [24] | N/A | 27 | 32 [24] | |
IBM | IBM Toronto [22] | Superconducting | Falcon r4 [24] | N/A | 27 | 32 [24] | |
Intel | 17-Qubit Superconducting Test Chip | Superconducting | 40-pin cross gap | N/A | 17 [25] [26] | October 10, 2017 | |
Intel | Tangle Lake | Superconducting | 108-pin cross gap | N/A | 49 [27] | January 9, 2018 | |
Intel | Tunnel Falls | Semiconductor spin qubits | 12 [28] | June 15, 2023 | |||
IonQ | Harmony | Trapped ion | All-to-All [24] | 99.73 (1 qubit) 90.02 (2 qubit) 99.30 (SPAM) | 11 [29] | 2022 | 8 [24] |
IonQ | Aria | Trapped ion | All-to-All [24] | 99.97 (1 qubit) 98.33 (2 qubit) 98.94 ((SPAM) | 25 [29] | 2022 | |
IonQ | Forte | Trapped ion | 366x1 chain [30] All-to-All [24] | 99.98 (1 qubit) 98.5–99.3 (2 qubit) [30] 99.56 ((SPAM) | 36 [29] (earlier 32) | 2022 | |
IQM | - | Superconducting | Star | 99.91 (1 qubit) 99.14 (2 qubits) | 5 [31] | November 30, 2021 [32] | N/A |
IQM | - | Superconducting | Square lattice | 99.91 (1 qubit median) 99.944 (1 qubit max) 98.25 (2 qubits median) 99.1 (2 qubits max) | 20 | October 9, 2023 [33] | 16 [34] |
M Squared Lasers | Maxwell | Neutral atoms in optical lattices | 99.5 (3-qubit gate), 99.1 (4-qubit gate) [35] | 200 [36] | November 2022 | ||
Oxford Quantum Circuits | Lucy [37] | Superconducting | 8 | 2022 | |||
Oxford Quantum Circuits | OQC Toshiko [38] | Superconducting | 32 | 2023 | |||
Quandela | Ascella | Photonics | N/A | 99.6 (1 qubit) 93.8 (2 qubits) 86.0 (3 qubits) | 6 [39] | 2022 [40] | |
QuTech at TU Delft | Spin-2 | Semiconductor spin qubits | 99 (average gate) 85 (readout) [41] | 2 | 2020 | ||
QuTech at TU Delft | - | Semiconductor spin qubits | 6 [42] | September 2022 | |||
QuTech at TU Delft | Starmon-5 | Superconducting | X configuration | 97 (readout) [43] | 5 | 2020 | |
Quantinuum | H2 [44] | Trapped ion | Racetrack, All-to-All | 99.997 (1 qubit) 99.87 (2 qubit) | 56 [45] (earlier 32) | May 9, 2023 | 2,097,152 [46] |
Quantinuum | H1-1 [47] | Trapped ion | 15×15 (Circuit Size) | 99.996 (1 qubit) 99.914 (2 qubit) | 20 | 2022 | 1,048,576 [48] |
Quantinuum | H1-2 [47] | Trapped ion | All-to-All [24] | 99.996 (1 qubit) 99.7 (2 qubit) | 12 | 2022 | 4096 [49] |
Quantware | Soprano [50] | Superconducting | 99.9 (single-qubit gates) | 5 | July 2021 | ||
Quantware | Contralto [51] | Superconducting | 99.9 (single-qubit gates) | 25 | March 7, 2022 [52] | ||
Quantware | Tenor [53] | Superconducting | 64 | February 23, 2023 | |||
Rigetti | Agave | Superconducting | N/A | 96 (Single-qubit gates) 87 (Two-qubit gates) | 8 | June 4, 2018 [54] | |
Rigetti | Acorn | Superconducting transmon | N/A | 98.63 (Single-qubit gates) 87.5 (Two-qubit gates) | 19 [55] | December 17, 2017 | |
Rigetti | Aspen-1 | Superconducting | N/A | 93.23 (Single-qubit gates) 90.84 (Two-qubit gates) | 16 | November 30, 2018 [54] | |
Rigetti | Aspen-4 | Superconducting | 99.88 (Single-qubit gates) 94.42 (Two-qubit gates) | 13 | March 10, 2019 | ||
Rigetti | Aspen-7 | Superconducting | 99.23 (Single-qubit gates) 95.2 (Two-qubit gates) | 28 | November 15, 2019 | ||
Rigetti | Aspen-8 | Superconducting | 99.22 (Single-qubit gates) 94.34 (Two-qubit gates) | 31 | May 5, 2020 | ||
Rigetti | Aspen-9 | Superconducting | 99.39 (Single-qubit gates) 94.28 (Two-qubit gates) | 32 | February 6, 2021 | ||
Rigetti | Aspen-10 | Superconducting | 99.37 (Single-qubit gates) 94.66 (Two-qubit gates) | 32 | November 4, 2021 | ||
Rigetti | Aspen-11 | Superconducting | Octagonal [24] | 99.8 (Single-qubit gates) 92.7 (Two-qubit gates CZ) 91.0 (Two-qubit gates XY) | 40 | December 15, 2021 | |
Rigetti | Aspen-M-1 | Superconducting transmon | Octagonal [24] | 99.8 (Single-qubit gates) 93.7 (Two-qubit gates CZ) 94.6 (Two-qubit gates XY) | 80 | February 15, 2022 | 8 [24] |
Rigetti | Aspen-M-2 | Superconducting transmon | 99.8 (Single-qubit gates) 91.3 (Two-qubit gates CZ) 90.0 (Two-qubit gates XY) | 80 | August 1, 2022 | ||
Rigetti | Aspen-M-3 | Superconducting transmon | N/A | 99.9 (Single-qubit gates) 94.7 (Two-qubit gates CZ) 95.1 (Two-qubit gates XY) | 80 [56] | December 2, 2022 | |
Rigetti | Ankaa-2 | Superconducting transmon | N/A | 98 (Two-qubit gates) | 84 [57] | December 20, 2023 | |
RIKEN | RIKEN [58] | Superconducting | N/A | N/A | 53 effective (64 total) [59] [60] | March 27, 2023 | N/A |
SaxonQ | Princess | Nitrogen-vacancy center | 4 [61] | June 26, 2024 | |||
SpinQ | Triangulum | Nuclear magnetic resonance | 3 [62] | September 2021 | |||
USTC | Jiuzhang | Photonics | N/A | N/A | 76 [63] [64] | 2020 | |
USTC | Zuchongzhi | Superconducting | N/A | N/A | 62 [65] | 2020 | |
USTC | Zuchongzhi 2.1 | Superconducting | lattice [66] | 99.86 (Single-qubit gates) 99.41 (Two-qubit gates) 95.48 (Readout) | 66 [67] | 2021 | |
USTC | Zuchongzhi 3.0 [68] | Superconducting transmon | 15 x 7 | 99.90 (Single-qubit gates) 99.62 (Two-qubit gates) 99.18 (Readout) | 105 | December 16, 2024 | |
Xanadu | Borealis [69] | Photonics (Continuous-variable) | N/A | N/A | 216 [69] | 2022 [69] | |
Xanadu | X8 [70] | Photonics (Continuous-variable) | N/A | N/A | 8 | 2020 | |
Xanadu | X12 | Photonics (Continuous-variable) | N/A | N/A | 12 | 2020 [70] | |
Xanadu | X24 | Photonics (Continuous-variable) | N/A | N/A | 24 | 2020 [70] | |
CAS | Xiaohong [71] | Superconducting | N/A | N/A | 504 [71] | 2024 |
These QPUs are based on quantum annealing, not to be confused with digital annealing. [72]
Manufacturer | Name/Codename /Designation | Architecture | Layout | Fidelity (%) | Qubits | Release date |
---|---|---|---|---|---|---|
D-Wave | D-Wave One (Rainier) | Superconducting | C4 = Chimera(4,4,4) [73] = 4×4 K4,4 | N/A | 128 | May 11, 2011 |
D-Wave | D-Wave Two | Superconducting | C8 = Chimera(8,8,4) [73] = 8×8 K4,4 | N/A | 512 | 2013 |
D-Wave | D-Wave 2X | Superconducting | C12 = Chimera(12,12,4) [73] = 12×12 K4,4 | N/A | 1152 | 2015 |
D-Wave | D-Wave 2000Q | Superconducting | C16 = Chimera(16,16,4) [73] = 16×16 K4,4 | N/A | 2048 | 2017 |
D-Wave | D-Wave Advantage | Superconducting | Pegasus P16 [74] | N/A | 5760 | 2020 |
D-Wave | D-Wave Advantage 2 [75] [76] [77] [78] | Superconducting [75] [76] | Zephyr Z15 [78] [79] | N/A | 7440 [80] | 2024 [75] [76] [77] [78] [79] |
These QPUs are based on analog Hamiltonian simulation.
Manufacturer | Name/Codename/Designation | Architecture | Layout | Fidelity (%) | Qubits | Release date |
---|---|---|---|---|---|---|
QuEra | Aquila | Neutral atoms | N/A | N/A | 256 [81] | November 2022 |
A quantum computer is a computer that exploits quantum mechanical phenomena. On small scales, physical matter exhibits properties of both particles and waves, and quantum computing leverages this behavior using specialized hardware. Classical physics cannot explain the operation of these quantum devices, and a scalable quantum computer could perform some calculations exponentially faster than any modern "classical" computer. Theoretically a large-scale quantum computer could break some widely used encryption schemes and aid physicists in performing physical simulations; however, the current state of the art is largely experimental and impractical, with several obstacles to useful applications.
This is a timeline of quantum computing.
In logic circuits, the Toffoli gate, also known as the CCNOT gate (“controlled-controlled-not”), invented by Tommaso Toffoli, is a CNOT gate with two control qubits and one target qubit. That is, the target qubit will be inverted if the first and second qubits are both 1. It is a universal reversible logic gate, which means that any classical reversible circuit can be constructed from Toffoli gates.
Superconducting quantum computing is a branch of solid state physics and quantum computing that implements superconducting electronic circuits using superconducting qubits as artificial atoms, or quantum dots. For superconducting qubits, the two logic states are the ground state and the excited state, denoted respectively. Research in superconducting quantum computing is conducted by companies such as Google, IBM, IMEC, BBN Technologies, Rigetti, and Intel. Many recently developed QPUs use superconducting architecture.
Quantum programming is the process of designing or assembling sequences of instructions, called quantum circuits, using gates, switches, and operators to manipulate a quantum system for a desired outcome or results of a given experiment. Quantum circuit algorithms can be implemented on integrated circuits, conducted with instrumentation, or written in a programming language for use with a quantum computer or a quantum processor.
Quantum annealing (QA) is an optimization process for finding the global minimum of a given objective function over a given set of candidate solutions, by a process using quantum fluctuations. Quantum annealing is used mainly for problems where the search space is discrete with many local minima; such as finding the ground state of a spin glass or solving the traveling salesman problem. The term "quantum annealing" was first proposed in 1988 by B. Apolloni, N. Cesa Bianchi and D. De Falco as a quantum-inspired classical algorithm. It was formulated in its present form by T. Kadowaki and H. Nishimori in 1998, though an imaginary-time variant without quantum coherence had been discussed by A. B. Finnila, M. A. Gomez, C. Sebenik and J. D. Doll in 1994.
D-Wave Quantum Systems Inc. is a quantum computing company with locations in Palo Alto, California and Burnaby, British Columbia. D-Wave claims to be the world's first company to sell computers that exploit quantum effects in their operation. D-Wave's early customers include Lockheed Martin, the University of Southern California, Google/NASA, and Los Alamos National Laboratory.
D-Wave Two is the second commercially available quantum computer, and the successor to the first commercially available quantum computer, D-Wave One. Both computers were developed by Canadian company D-Wave Systems. The computers are not general purpose, but rather are designed for quantum annealing. Specifically, the computers are designed to use quantum annealing to solve a single type of problem known as quadratic unconstrained binary optimization. As of 2015, it was still debated whether large-scale entanglement takes place in D-Wave Two, and whether current or future generations of D-Wave computers will have any advantage over classical computers.
Quantum machine learning is the integration of quantum algorithms within machine learning programs.
The USC-Lockheed Martin Quantum Computing Center (QCC) is a joint scientific research effort between Lockheed Martin Corporation and the University of Southern California (USC). The QCC is housed at the Information Sciences Institute (ISI), a computer science and engineering research unit of the USC Viterbi School of Engineering, and is jointly operated by ISI and Lockheed Martin.
IBM Quantum Platform is an online platform allowing public and premium access to cloud-based quantum computing services provided by IBM. This includes access to a set of IBM's prototype quantum processors, a set of tutorials on quantum computation, and access to an interactive textbook. As of February 2021, there are over 20 devices on the service, six of which are freely available for the public. This service can be used to run algorithms and experiments, and explore tutorials and simulations around what might be possible with quantum computing.
Cloud-based quantum computing is the invocation of quantum emulators, simulators or processors through the cloud. Increasingly, cloud services are being looked on as the method for providing access to quantum processing. Quantum computers achieve their massive computing power by initiating quantum physics into processing power and when users are allowed access to these quantum-powered computers through the internet it is known as quantum computing within the cloud.
In quantum computing, quantum supremacy or quantum advantage is the goal of demonstrating that a programmable quantum computer can solve a problem that no classical computer can solve in any feasible amount of time, irrespective of the usefulness of the problem. The term was coined by John Preskill in 2011, but the concept dates to Yuri Manin's 1980 and Richard Feynman's 1981 proposals of quantum computing.
Rigetti Computing, Inc. is a Berkeley, California-based developer of quantum integrated circuits used for quantum computers. Rigetti also develops a cloud platform called Forest that enables programmers to write quantum algorithms.
Quil is a quantum instruction set architecture that first introduced a shared quantum/classical memory model. It was introduced by Robert Smith, Michael Curtis, and William Zeng in A Practical Quantum Instruction Set Architecture. Many quantum algorithms require a shared memory architecture. Quil is being developed for the superconducting quantum processors developed by Rigetti Computing through the Forest quantum programming API. A Python library called pyQuil
was introduced to develop Quil programs with higher level constructs. A Quil backend is also supported by other quantum programming environments.
In quantum computing, a qubit is a unit of information analogous to a bit in classical computing, but it is affected by quantum mechanical properties such as superposition and entanglement which allow qubits to be in some ways more powerful than classical bits for some tasks. Qubits are used in quantum circuits and quantum algorithms composed of quantum logic gates to solve computational problems, where they are used for input/output and intermediate computations.
Quantum volume is a metric that measures the capabilities and error rates of a quantum computer. It expresses the maximum size of square quantum circuits that can be implemented successfully by the computer. The form of the circuits is independent from the quantum computer architecture, but compiler can transform and optimize it to take advantage of the computer's features. Thus, quantum volumes for different architectures can be compared.
Quantinuum is a quantum computing company formed by the merger of Cambridge Quantum and Honeywell Quantum Solutions. The company's H-Series trapped-ion quantum computers set the highest quantum volume to date of 1,048,576 in April 2024. This architecture supports all-to-all qubit connectivity, allowing entangled states to be created between all qubits, and enables a high fidelity of quantum states.
The current state of quantum computing is referred to as the noisy intermediate-scale quantum (NISQ) era, characterized by quantum processors containing up to 1,000 qubits which are not advanced enough yet for fault-tolerance or large enough to achieve quantum advantage. These processors, which are sensitive to their environment (noisy) and prone to quantum decoherence, are not yet capable of continuous quantum error correction. This intermediate-scale is defined by the quantum volume, which is based on the moderate number of qubits and gate fidelity. The term NISQ was coined by John Preskill in 2018.
This glossary of quantum computing is a list of definitions of terms and concepts used in quantum computing, its sub-disciplines, and related fields.
{{cite web}}
: CS1 maint: numeric names: authors list (link)Advantage 2™ quantum system will incorporate a new qubit design that enables 20-way connectivity in a new topology. The Advantage 2 QPU will contain 7000+ qubits and make use of the latest improvements in quantum coherence in a multi-layer fabrication stack, further harnessing the quantum mechanical power of the system for finding better solutions, faster.