Skip to content
ContentLora

    Tip: press / anywhere to search.

    technology

    Superconducting qubits

    Also known as superconducting circuits, transmon qubits

    Superconducting qubits are tiny electrical circuits on chips that behave as qubits when cooled to millikelvin temperatures in a dilution refrigerator.[1] They power Google's Willow, IBM's Nighthawk and China's Zuchongzhi 3.0, and hosted the first below-threshold error correction in 2024.[2][3][4][5]

    Editor reviewedUpdated Quantum computingComputingPhysics
    Key facts

    How they work

    A superconducting qubit is an electrical circuit patterned on a chip that behaves as a qubit when cooled to near absolute zero. The circuits are kept at roughly 15 to 20 millikelvin inside a dilution refrigerator; IBM’s first cloud-accessible quantum computer used five such qubits in 2016.[1] Because they are made with chip-fabrication techniques, companies run dedicated fabs: Google built Willow in its Santa Barbara facility, and IBM moved primary fabrication to 300 mm wafers at the Albany NanoTech Complex.[6][7] IBM says that move doubled its development speed and let it make chips ten times more physically complex.[8] For how conventional chips are made at scale, see how chips are made.

    Strengths and weaknesses

    Superconducting gates are fast: Google’s surface code ran an error-correction cycle every 1.1 microseconds.[9] The trade-off is short coherence; Willow’s T1 times approach 100 microseconds, a five-fold improvement on Google’s previous chips.[10] By comparison, neutral atoms have held superposition for about 13 seconds.[11] Qubits are typically wired to neighbours on a square lattice, the layout the surface-code is designed for.[12] Every coupler needs control wiring, so Google has tested error correction on a hexagonal pattern in which each qubit links to three neighbours instead of four, which it says would simplify chip design and fabrication.[13] Superconducting circuits are analog devices that drift, so their control settings must be recalibrated; Google reported in 2026 that a reinforcement-learning agent could retune thousands of parameters while a computation runs.[14] Two-qubit gate fidelities trail trapped ions: Zuchongzhi 3.0 reported 99.62%, against 99.921% on Quantinuum’s Helios.[15][16]

    Landmark results

    • Sycamore (2019). Google reported that its 54-qubit Sycamore chip performed a computation in 200 seconds that it estimated would take the fastest supercomputer 10,000 years, the first quantum advantage claim.[17][18]
    • Below threshold (2024). Google’s Willow halved logical errors each time its surface code grew from 3x3 to 7x7.[5]
    • Random circuit sampling. Willow and USTC’s Zuchongzhi 3.0 each reported sampling tasks far beyond classical supercomputers.[19][4] Google calls random circuit sampling the classically hardest benchmark a quantum computer can run today.[20]
    • Verifiable advantage (2025). Google’s Quantum Echoes ran 13,000 times faster than Frontier on a physics observable.[21]
    • Advantage claims (2026). IBM and partners claimed quantum advantage on IBM processors in July 2026; a preprint later reproduced the IBM-University of Chicago experiment classically.[22][23][24]

    Roadmaps

    IBM plans the fault-tolerant Starling machine with 200 logical qubits by 2029, built from qLDPC-coded modules.[25][26] Before then it expects Nighthawk-class processors to grow from 5,000 two-qubit gates to 7,500 by the end of 2026 and 15,000 by 2028.[27] Google’s stated next goal is a useful, beyond-classical computation on a real-world problem.[28] IBM is one of four teams DARPA moved into the final stage of its Quantum Benchmarking Initiative in October 2026, listed for its modular superconducting processors, and Canada’s Nord Quantique, which uses superconducting qubits with bosonic error correction, reached Stage B.[29][30] Microsoft’s topological design, also in Stage C, is described by DARPA as a superconducting topological architecture.[31]

    Questions readers ask

    Why do superconducting qubits need such cold temperatures?

    They must be kept at roughly 15 to 20 millikelvin in a dilution refrigerator to stay in a stable quantum state.[1]

    How long do superconducting qubits keep their state?

    Google's Willow qubits have T1 coherence times approaching 100 microseconds.[10]

    Who builds superconducting quantum computers?

    Leading examples include Google (Willow), IBM (Nighthawk and Loon) and the University of Science and Technology of China (Zuchongzhi 3.0).[2][3][4]

    Sources

    Each numbered claim is a statement we checked against the sources listed with it. Status shows how well established it is.

    1. [1]

      IBM put the first quantum computer on the cloud on 4 May 2016, a five-qubit superconducting device kept in a dilution refrigerator at roughly 15 to 20 millikelvin. confirmedas of 2021-05-04

    2. [2]

      Google's Willow chip, announced on 9 December 2024, has 105 superconducting qubits. confirmedas of 2024-12-09

    3. [3]

      IBM unveiled the Nighthawk processor in November 2025, with 120 qubits linked by 218 tunable couplers, allowing circuits about 30 percent more complex than its previous processors. confirmedas of 2025-11-12

    4. [4]

      The University of Science and Technology of China reported in Physical Review Letters in March 2025 that its 105-qubit Zuchongzhi 3.0 superconducting processor ran an 83-qubit, 32-layer random circuit sampling task it estimated at 10^15 times faster than the most powerful supercomputer. confirmedas of 2025-03-06

    5. [5]

      Google reported that as it scaled Willow's encoded qubit grids from 3x3 to 5x5 to 7x7, the logical error rate halved at each step, which it called the first below-threshold error correction. confirmedas of 2024-12-09

    6. [6]

      Willow was fabricated in Google's dedicated quantum chip fabrication facility in Santa Barbara, California. confirmedas of 2024-12-09

    7. [7]

      IBM moved primary quantum chip fabrication to a 300 mm wafer facility at the Albany NanoTech Complex in New York. confirmedas of 2025-11-12

    8. [8]

      IBM said the move to 300 mm wafer fabrication doubled its development speed while raising the physical complexity of its quantum chips tenfold. confirmedas of 2025-11-12

    9. [9]

      The experiment decoded errors in real time, with an average decoder latency of 63 microseconds at distance 5 and an error-correction cycle time of 1.1 microseconds. confirmedas of 2025-02-01

    10. [10]

      Willow's qubits have T1 coherence times approaching 100 microseconds, about five times longer than Google's previous generation. confirmedas of 2024-12-09

    11. [11]

      In September 2025 a Caltech team reported in Nature an array of 6,100 cesium-atom qubits held by optical tweezers (one laser split into 12,000 tweezers), with superposition lasting about 13 seconds and 99.98% accuracy in manipulating individual qubits; entangling the qubits for full computations was the stated next step. confirmedas of 2025-09-24

    12. [12]

      In the surface code, data qubits and measure qubits are interleaved on a square lattice; repeated measurements flag physical errors without disturbing the stored logical information. confirmedas of 2026-01-13

    13. [13]

      On a hexagonal lattice each qubit would connect to three neighbours instead of four, which Google says would simplify the design and fabrication of large chips; it tested this on Willow by switching off unused couplers. confirmedas of 2026-01-13

    14. [14]

      Google's July 2026 Nature paper describes a reinforcement-learning agent that learns from error-detection data to steer thousands of control parameters during a computation, instead of halting the computation to recalibrate. confirmedas of 2026-07-22

    15. [15]

      Zuchongzhi 3.0 reported 72-microsecond coherence, 99.90% single-qubit and 99.62% two-qubit gate fidelity. confirmedas of 2025-03-06

    16. [16]

      Quantinuum reports Helios two-qubit gate fidelity of 99.921% and single-qubit gate fidelity of 99.9975%, with all 98 qubits fully connected. confirmedas of 2026-10-10

    17. [17]

      On 23 October 2019 Google reported that its 54-qubit Sycamore processor performed a target computation in 200 seconds that it estimated would take the world's fastest supercomputer 10,000 years. confirmedas of 2019-10-23

    18. [18]

      Quantum computational advantage was first claimed in 2019, and later experiments have reinforced the claim. confirmedas of 2026-09-08

    19. [19]

      Google said Willow performed a random circuit sampling computation in under five minutes that would take one of the fastest supercomputers 10 septillion years. confirmedas of 2024-12-09

    20. [20]

      Google describes random circuit sampling, which it pioneered, as the classically hardest benchmark that can be run on a quantum computer today and an entry-point test of beating classical computers. confirmedas of 2024-12-09

    21. [21]

      On 22 October 2025 Google reported in Nature that its Quantum Echoes (out-of-time-order correlator) algorithm ran on Willow in about two hours, a task it estimated would take 13,000 times longer on the Frontier supercomputer. confirmedas of 2025-10-22

    22. [22]

      On 30 July 2026 IBM and Algorithmiq claimed quantum advantage for a simulation of heterogeneous quantum matter run on IBM's Heron processor, a problem posted on the quantum advantage tracker about eight months earlier. confirmedas of 2026-07-30

    23. [23]

      Also on 30 July 2026 IBM and the University of Chicago claimed quantum advantage using an error-detecting encoded circuit on 70 logical qubits, run in about 15 minutes, with effective logical error rates 10 times lower than physical error rates. confirmedas of 2026-07-30

    24. [24]

      An August 2026 preprint reported classically reproducing all 2,051 amplitude batches of IBM's 70-qubit doped Clifford sampling experiment in 37.3 minutes on 256 NVIDIA H100 GPUs, with results compatible with IBM's fidelity bound. reportedas of 2026-08-13

    25. [25]

      In June 2025 IBM said it would build IBM Quantum Starling, a fault-tolerant quantum computer with 200 logical qubits able to run 100 million quantum operations, in Poughkeepsie, New York, by 2029. confirmedas of 2025-06-10

    26. [26]

      IBM's fault-tolerance plan uses quantum low-density parity check (qLDPC) codes, which it says cut error-correction overhead by about 90 percent compared with other leading codes. confirmedas of 2025-06-10

    27. [27]

      IBM said Nighthawk supports circuits of up to 5,000 two-qubit gates and expected later versions to reach 7,500 gates by the end of 2026, 10,000 in 2027 and 15,000 by 2028 with 1,000 or more connected qubits. confirmedas of 2025-11-12· forecast

    28. [28]

      Google's stated next goal is a first useful, beyond-classical computation relevant to a real-world application, citing areas such as drug discovery, battery design and energy research. confirmedas of 2024-12-09

    29. [29]

      DARPA lists the October 2026 Stage C entrants as Atom Computing of Boulder, Colorado (neutral-atom arrays), Diraq of Sydney (silicon CMOS spin qubits), IBM of Yorktown Heights, New York (modular superconducting processors) and IonQ of College Park, Maryland (trapped ions). confirmedas of 2026-10-07

    30. [30]

      DARPA lists the QBI Stage B approaches as neutral atoms (Atom Computing, QuEra), trapped ions (IonQ, Quantinuum), superconducting circuits (IBM, and Nord Quantique with bosonic error correction), silicon spin or atom qubits (Diraq, Photonic, Quantum Motion, Silicon Quantum Computing) and photonics (Xanadu). confirmedas of 2025-11-06

    31. [31]

      DARPA said more than 50 experts from its test and evaluation team examined Microsoft's and PsiQuantum's approaches, and described Microsoft's design as a compact superconducting topological qubit architecture. confirmedas of 2025-02-06

    32. [32]

      Google's below-threshold experiment, published in Nature in 2025, ran a 101-qubit distance-7 surface code with 0.143% logical error per cycle, with errors suppressed by a factor of 2.14 each time the code distance grew by two. confirmedas of 2025-02-01

    33. [33]

      On 6 November 2025 DARPA moved 11 companies to QBI Stage B - Atom Computing, Diraq, IBM, IonQ, Nord Quantique, Photonic, Quantinuum, Quantum Motion, QuEra, Silicon Quantum Computing and Xanadu. confirmedas of 2025-11-06

    Revision history (2)
    1. Page created.
    2. Added the 2019 Sycamore milestone, IBM's 300 mm fabrication gains and gate-count roadmap, Google's hexagonal-lattice and drift-control work, and DARPA Stage C details.

    Created Oct 10, 2026. Last reviewed by an editor on Oct 10, 2026. Next scheduled review: Jan 10, 2027.

    Cite this page

    "Superconducting qubits." ContentLora, updated Oct 10, 2026. https://contentlora.com/wiki/superconducting-qubits

    Spotted an error? Suggest a correction or emailcorrections@contentlora.com.