QPU

A Quantum Processing Unit (QPU), also known as a quantum processor, is the core hardware component of a quantum computer that manipulates qubits using the principles of quantum mechanics to perform computations that are intractable for classical processors.^[1][2] Unlike classical CPUs that process bits in binary states (0 or 1), QPUs leverage quantum phenomena such as superposition and entanglement to process qubits that can exist in multiple states simultaneously, enabling exponential scaling of computational power for specific problems.^[3] By analogy with the graphics processing unit (GPU) and tensor processing unit (TPU), a QPU is a specialized accelerator dispatched by a classical host, not a general-purpose processor. A more detailed companion overview is kept at Quantum processor.

QPUs are designed to solve complex problems in cryptography, molecular simulation, optimization, and machine learning that would require impractical amounts of time on even the most powerful supercomputers.^[4][5] As of 2026, no QPU has delivered a broadly useful commercial advantage, but the field has passed several technical milestones: Google's 105-qubit Willow processor demonstrated below-threshold quantum error correction and ran a random circuit sampling benchmark in under five minutes that Google estimated would take a leading supercomputer roughly 10^25 (10 septillion) years,^[6] IBM has committed to delivering Starling, a fault-tolerant machine with 200 logical qubits, by 2029,^[7] and in October 2025 Google's Quantum Echoes experiment became the first verifiable quantum algorithm to run about 13,000 times faster than the best classical method on a leading supercomputer.^[26]

Overview

The QPU represents the quantum analog to the classical CPU, serving as the computational heart of quantum computers. However, this analogy can be misleading; QPUs are more accurately described as specialized co-processors similar to GPUs or TPUs, designed to accelerate specific computational tasks rather than serve as general-purpose processors.^[8]

How is a QPU different from a CPU or GPU?

A QPU operates by manipulating qubits through quantum logic gates and interactions, forming quantum circuits analogous to classical logic circuits. While classical processors execute operations sequentially or in limited parallelism, QPUs can explore an exponentially large computational space simultaneously: N qubits can represent 2^N states at once, compared to N bits representing only one of 2^N possible states.^[9] A CPU and a GPU both operate on definite bits and scale performance by adding cores or threads; a QPU instead scales the size of the state space it can represent, but only for the narrow class of problems whose structure a quantum algorithm can exploit. In practice a QPU never runs alone: a classical host handles compilation, error decoding, and measurement post-processing, so a quantum computer is a hybrid quantum-classical system.

Modern quantum computers integrate the QPU within complex systems including:

• Quantum chip: Physical substrate containing qubits and control elements

• Control electronics: Generates precisely timed microwave or laser pulses

• Cryogenic systems: Dilution refrigerators maintaining temperatures near absolute zero (10-15 millikelvin) for superconducting QPUs

• Classical processors: Handle compilation, error correction, and result processing

• Shielding: Electromagnetic and vibrational isolation protecting fragile quantum states^[2][10]

History

When was the quantum computer first proposed? Theoretical Foundations (1980-1985)

The theoretical foundations of quantum computing emerged in the early 1980s. Paul Benioff proposed quantum mechanical models of computation in 1980, describing a quantum mechanical Turing machine.^[11] Richard Feynman formalized the concept in his 1981 MIT keynote "Simulating Physics with Computers," arguing that quantum computers would be necessary to efficiently simulate quantum systems, and published these ideas in 1982.^[12] David Deutsch provided the theoretical framework for universal quantum computers in 1985, establishing that quantum systems could efficiently simulate any physical process.^[13]

Early Implementations (1990s)

The 1990s transformed quantum computing from theoretical concept to experimental reality:

• 1994: Peter Shor developed Shor's algorithm for integer factorization, demonstrating exponential quantum speedup for breaking RSA encryption^[14]

• 1995: Dave Wineland and Christopher Monroe demonstrated the first two-qubit quantum circuit at NIST^[15]

• 1996: Lov Grover developed Grover's algorithm providing quadratic speedup for unstructured search^[16]

• 1998: IBM researchers executed Grover's algorithm on a 2-qubit NMR quantum computer^[17]

• 1999: NEC demonstrated superconducting qubits using Josephson junctions, establishing the foundation for modern superconducting QPUs^[18]

Commercial Development (2000s-2010s)

Scroll sideways for more →

Modern Era (2020s)

The 2020s have witnessed rapid acceleration in QPU development:

• 2021: IBM's Eagle processor achieved 127 qubits, surpassing the 100-qubit barrier^[20]

• 2022: IBM Osprey reached 433 qubits; Xanadu demonstrated photonic quantum advantage with Borealis^[21]

• 2023: IBM Condor achieved 1,121 qubits; Atom Computing demonstrated 1,180 neutral atom qubits^[22][23]

• 2024: Google's Willow achieved below-threshold error correction; Quantinuum reached 99.914% two-qubit fidelity; Microsoft and Atom Computing entangled 24 logical qubits, the most on record at the time^[6][24][51]

• 2025: Microsoft unveiled the Majorana 1 topological qubit prototype (8 qubits); PsiQuantum announced Omega, a manufacturable photonic chipset; Craig Gidney of Google cut the estimated cost of breaking RSA-2048 to under one million noisy qubits; Google reported Quantum Echoes, a verifiable technique that enabled a ~13,000x speed-up for a molecular-structure task; Quantinuum launched the 98-qubit Helios; IonQ reached #AQ 64; and NVIDIA introduced NVQLink to couple QPUs with GPU supercomputers^{[25][26][44][45][46][47][48][49]}

Fundamental Principles

QPUs derive their computational power from three core quantum mechanical phenomena that have no classical analog:

Superposition

Quantum superposition allows qubits to exist in a probabilistic combination of both |0⟩ and |1⟩ states simultaneously, described mathematically as:

where α and β are complex probability amplitudes satisfying |α|² + |β|² = 1.^[9]

This enables N qubits to represent all 2^N possible states simultaneously, providing exponential scaling unachievable classically. For perspective, 300 qubits can theoretically represent more states (2^300, roughly 10^90) than there are atoms in the observable universe (about 10^80).^[1]

Entanglement

Quantum entanglement creates correlations between qubits where the quantum state of one qubit cannot be described independently of others. A classic example is the Bell state:

Entangled qubits exhibit perfect correlations regardless of physical separation; measuring one instantly determines the state of others, enabling quantum algorithms to process information in ways impossible classically.^[9][27]

Interference

Quantum interference manipulates probability amplitudes to constructively amplify correct answers while destructively canceling incorrect ones. Quantum algorithms carefully orchestrate interference patterns so that paths leading to wrong answers cancel out while correct solutions reinforce, dramatically increasing the probability of measuring the desired result.^[27]

Physical Implementations

Multiple competing technologies exist for implementing QPUs, each with distinct advantages and challenges:

Comparison of Leading Qubit Technologies

Scroll sideways for more →

Superconducting QPUs

Superconducting QPUs dominate current commercial systems, using microscopic circuits with Josephson junctions cooled to near absolute zero where they exhibit quantum behavior.^[28]

• Advantages: Fast gate operations (10-100 ns), leverages semiconductor fabrication, demonstrated scaling to >1,000 qubits

• Challenges: Short coherence times (30-300 μs), requires extreme cooling, crosstalk between qubits

• Notable Systems: IBM Condor (1,121 qubits), Google Willow (105 qubits with 99.67% two-qubit fidelity and T1 coherence approaching 100 μs), Rigetti Ankaa-3 (84 qubits), IBM Nighthawk (120 qubits, up to 5,000 two-qubit gates, 2025)^[6][29][45]

Trapped-Ion QPUs

Trapped-ion systems suspend individual charged atoms in electromagnetic fields, manipulating them with precision lasers.^[30]

• Advantages: Exceptional coherence (seconds to minutes), highest gate fidelities (>99.9%), all-to-all connectivity

• Challenges: Slower gates (10-100 μs), complex scaling beyond 50-100 ions

• Notable Systems: Quantinuum Helios (98 qubits, 99.921% two-qubit fidelity, 48 error-corrected logical qubits, 2025), Quantinuum H2 (56 qubits with quantum volume of 8,388,608), IonQ Tempo (100 qubits, #AQ 64)^{[31][32][46][47]}

Photonic QPUs

Photonic processors encode quantum information in properties of light particles (photons).^[33]

• Advantages: Room temperature operation, minimal decoherence, compatible with fiber networks

• Challenges: Difficult two-qubit gates, photon loss, probabilistic operations

• Notable Systems: Xanadu Borealis (216 modes), PsiQuantum Omega (manufacturable chipset fabricated at GlobalFoundries, with a reported two-qubit fusion gate fidelity of 99.22%, 2025)^[49]

Neutral-Atom QPUs

This approach uses arrays of neutral atoms trapped by focused laser beams ("optical tweezers").^[34]

• Advantages: High scalability (>1,000 qubits demonstrated), flexible connectivity, long coherence

• Challenges: Atom loss during computation, moderate gate speeds

• Notable Systems: Atom Computing (1,180 qubits), QuEra Aquila (256 qubits), Pasqal (analog neutral-atom systems)^[35]

Topological QPUs

Microsoft's approach uses exotic quasiparticles called Majorana zero modes in specially engineered materials.^[25]

• Advantages: Theoretical intrinsic error protection, potential for million-qubit chips

• Challenges: Unproven at scale, limited experimental validation

• Notable Systems: Microsoft Majorana 1 (8-qubit prototype, early stage). The accompanying 2025 Nature paper demonstrated single-shot parity measurement in an indium arsenide-aluminum device, though the Nature editors noted the results do not by themselves establish the presence of Majorana zero modes, and parts of the physics community remain skeptical^[25]

Architecture and Operation

System Components

Modern QPUs integrate multiple subsystems working in concert:

Scroll sideways for more →

Operational Workflow

The execution of a quantum algorithm follows these steps:

1. Compilation: Classical compiler translates high-level algorithm into hardware-specific quantum circuits

2. Initialization: All qubits prepared in known state (typically |0⟩)

3. Gate Sequence: Control pulses apply quantum gates according to compiled circuit

4. Measurement: Qubit states collapsed and read out as classical bits

5. Repetition: Process repeated thousands of times to build probability distribution

6. Post-processing: Classical analysis extracts final answer from measurement statistics^[1]

Quantum Gates

Quantum logic gates manipulate qubit states through controlled interactions:

• Single-qubit gates: Rotations around Bloch sphere axes (X, Y, Z rotations, Hadamard)

• Two-qubit gates: Create entanglement (CNOT, CZ, iSWAP)

• Gate implementation: Microwave pulses (superconducting), laser pulses (trapped ion/neutral atom)

• Fidelity requirements: >99.9% for practical error correction^[36]

Performance Metrics

Quantum Volume

Quantum volume combines multiple performance aspects into a single metric:^[37]

• Accounts for qubit count, connectivity, gate fidelity, and measurement accuracy

• Higher values indicate better overall system performance

• Record: Quantinuum H2 achieved quantum volume of 8,388,608 (2^23) in 2023^[32]

Key Performance Indicators

Scroll sideways for more →

Applications

QPUs excel at problems with inherent quantum structure or exponential classical complexity:

Quantum Simulation

• Drug Discovery: Simulating molecular interactions, protein folding, drug-target binding^[38]

• Materials Science: Designing catalysts, batteries, superconductors

• Example: In 2025, Google reported Quantum Echoes, a verifiable technique that enabled a ~13,000x speed-up for a molecular-structure task on the Willow chip, with a method for certifying the result's correctness and potential applications in nuclear magnetic resonance spectroscopy^[26]

Optimization

• Finance: Portfolio optimization, risk analysis, derivative pricing

• Logistics: Route optimization, supply chain, scheduling

• Manufacturing: Process optimization, quality control

• Example: In March 2025, researchers from JPMorganChase, Quantinuum, and two US national laboratories demonstrated certified quantum randomness on Quantinuum's 56-qubit H2 hardware, publishing the result in Nature; certified randomness is useful for Monte Carlo methods, cryptography, and fairness applications^[39]

Cryptography

• Code Breaking: Shor's algorithm threatens current RSA/ECC encryption

• Quantum Security: Quantum key distribution for unbreakable communication

• Resource Requirements: Breaking RSA-2048 with Shor's algorithm requires thousands of logical qubits. A 2019 analysis by Gidney and Ekera estimated roughly 20 million noisy physical qubits to factor RSA-2048 in about 8 hours; in May 2025, Craig Gidney published a revised estimate of fewer than one million noisy qubits running for under a week, a roughly 20x reduction that shortens the assumed runway for post-quantum cryptography while still leaving calendar timelines speculative^[40][44]

Machine Learning

• Quantum Machine Learning: Enhanced feature spaces, quantum neural networks

• Speedups: HHL algorithm for linear systems, quantum kernel methods

• Applications: Pattern recognition, optimization, data analysis^[41]

• Reality check: Most quantum machine learning algorithms are heuristic, and as of 2026 there is no proven end-to-end exponential quantum speedup for practical machine learning workloads; a widely cited 2023 analysis by Hoefler, Haner, and Troyer argued that quadratic speedups alone are unlikely to yield practical quantum advantage in the foreseeable future because of the enormous constant-factor and error-correction overheads^[52]

How do QPUs relate to AI and machine learning?

The relationship between QPUs and artificial intelligence runs in both directions, and the two fields increasingly share infrastructure.

AI for quantum. Deep learning is being used to make QPUs work better. Google DeepMind and Google Quantum AI built AlphaQubit, a neural network decoder for quantum error correction, and reported in Nature in November 2024 that it identified errors more accurately than the leading conventional decoders on data from a real superconducting processor.^[50] Following the naming lineage of AlphaGo and AlphaFold, AlphaQubit applies a learned model to a task, decoding, that had long been handled by hand-built algorithms. Other groups have applied reinforcement learning and generative models to discover error-correcting codes and tune control pulses.

Quantum for AI. QPUs may in principle accelerate parts of machine learning through quantum feature maps, kernel methods, and variational circuits, but as noted above these approaches remain heuristic and unproven at scale.^[52] A newer framing, which Quantinuum calls Generative Quantum AI (GenQAI), uses a QPU to generate high-quality training data (for example from quantum chemistry or magnetism simulations) that is then fed to classical AI models; Quantinuum positioned its 2025 Helios system around this idea.^[46]

Shared infrastructure. Because a QPU always runs alongside classical accelerators, hardware vendors are now tying QPUs directly to the GPU systems used for AI. In October 2025 NVIDIA introduced NVQLink, an open architecture that couples QPUs with GPU supercomputers at microsecond latency for real-time quantum error correction, integrated with its CUDA-Q software; NVIDIA said the launch involved 17 quantum hardware builders and nine US national laboratories.^[48]

Challenges and Limitations

Decoherence and Error Rates

• Decoherence: Quantum states decay in microseconds to seconds depending on technology

• Error accumulation: Current error rates of 0.1-1% per gate limit circuit depth

• Environmental sensitivity: Vibrations, electromagnetic fields, cosmic rays cause errors

• Mitigation strategies: Error correction codes, improved materials, better isolation^[42]

Quantum Error Correction

Quantum error correction enables fault-tolerant computation but requires massive overhead:

• Physical-to-logical ratio: 1,000-10,000 physical qubits per logical qubit with current error rates

• Surface code: Leading approach arranges qubits in 2D grid for error detection/correction

• Breakthrough: Google's Willow achieved below-threshold correction: the encoded error rate is cut roughly in half at each step from a 3x3 to a 5x5 to a 7x7 grid of physical qubits, so errors decrease as more qubits are added^[6]

• AI decoders: Neural-network decoders such as AlphaQubit now outperform classical decoders on real hardware data, and IBM's 2025 Loon processor demonstrated the components needed for quantum LDPC error correction^[50][45]

• Alternative codes: Quantum LDPC codes promise roughly 10x efficiency improvement over the surface code

Scalability Challenges

• Wiring complexity: Each qubit needs control/readout lines

• Cooling limits: Dilution refrigerators struggle beyond 10,000 qubits

• Control electronics: Classical processing requirements scale with qubit count

• Solutions: Modular architectures, cryogenic electronics, photonic interconnects

Current State and Future Outlook

NISQ Era (2025-2029)

The current Noisy Intermediate-Scale Quantum (NISQ) era, a term coined by physicist John Preskill in 2018, features:^[42]

• 50-1,000 noisy physical qubits

• Limited to ~1,000 gate operations before decoherence

• Specialized quantum advantages on narrow problems

• Hybrid quantum-classical algorithms (VQE, QAOA) showing practical value

Path to Fault Tolerance (2029-2035)

Industry roadmaps converge on achieving practical fault-tolerant systems:

Scroll sideways for more →

Market Projections

• Current market: roughly $1-2 billion annually (2025)

• 2030-2035 projection: analysts such as McKinsey project quantum technology revenue on the order of tens of billions of dollars, with wide uncertainty bands reflecting the unproven state of commercial advantage^[43]

• Key drivers: Drug discovery, materials science, financial modeling, logistics optimization^[43]

See Also

• Quantum computing

• Qubit

• Quantum algorithm

• Quantum error correction

• Quantum supremacy

• Quantum processor

• Quantum machine learning

• AlphaQubit

• Dilution refrigerator

• Post-quantum cryptography

References

1. IBM, "What is a quantum processing unit (QPU)?" IBM Think Topics. https://www.ibm.com/think/topics/qpu ↩
2. NVIDIA, "What Is a QPU?" NVIDIA Glossary. https://www.nvidia.com/en-us/glossary/quantum-processing-unit-qpu/ ↩
3. IBM Quantum, "What is quantum computing?" IBM. https://www.ibm.com/think/topics/quantum-computing ↩
4. Arute, F., et al., "Quantum supremacy using a programmable superconducting processor," Nature 574, 505-510 (2019). https://www.nature.com/articles/s41586-019-1666-5 ↩
5. Nielsen, M. A., and Chuang, I. L., Quantum Computation and Quantum Information, 10th Anniversary Edition, Cambridge University Press (2010). ↩
6. Google Quantum AI (Acharya, R., et al.), "Quantum error correction below the surface code threshold," Nature 638, 920-926 (2025). https://www.nature.com/articles/s41586-024-08449-y ↩
7. IBM, "IBM lays out clear path to fault-tolerant quantum computing," IBM Quantum blog (2025). https://www.ibm.com/quantum/blog/large-scale-ftqc ↩
8. NVIDIA, "What is quantum computing?" NVIDIA blog. https://blogs.nvidia.com/blog/what-is-quantum-computing/ ↩
9. Preskill, J., "Quantum computation lecture notes," Caltech Ph219. http://theory.caltech.edu/~preskill/ph229/ ↩
10. IBM Quantum, "The hardware and software for the era of quantum utility," IBM Quantum documentation. https://www.ibm.com/quantum ↩
11. Benioff, P., "The computer as a physical system: A microscopic quantum mechanical Hamiltonian model of computers as represented by Turing machines," Journal of Statistical Physics 22, 563-591 (1980). ↩
12. Feynman, R. P., "Simulating physics with computers," International Journal of Theoretical Physics 21, 467-488 (1982). https://doi.org/10.1007/BF02650179 ↩
13. Deutsch, D., "Quantum theory, the Church-Turing principle and the universal quantum computer," Proceedings of the Royal Society A 400, 97-117 (1985). ↩
14. Shor, P. W., "Algorithms for quantum computation: discrete logarithms and factoring," Proceedings 35th Annual Symposium on Foundations of Computer Science, 124-134 (1994). ↩
15. Monroe, C., Meekhof, D. M., King, B. E., Itano, W. M., and Wineland, D. J., "Demonstration of a fundamental quantum logic gate," Physical Review Letters 75, 4714-4717 (1995). ↩
16. Grover, L. K., "A fast quantum mechanical algorithm for database search," Proceedings 28th Annual ACM Symposium on Theory of Computing, 212-219 (1996). ↩
17. Chuang, I. L., Gershenfeld, N., and Kubinec, M., "Experimental implementation of fast quantum searching," Physical Review Letters 80, 3408-3411 (1998). ↩
18. Nakamura, Y., Pashkin, Y. A., and Tsai, J. S., "Coherent control of macroscopic quantum states in a single-Cooper-pair box," Nature 398, 786-788 (1999). ↩
19. Lockheed Martin / D-Wave Systems, "Lockheed Martin and D-Wave Systems sign multi-year quantum computing contract" (2011). https://www.dwavequantum.com/company/newsroom/ ↩
20. IBM, "IBM Quantum breaks the 100-qubit processor barrier," IBM Quantum blog (2021). https://www.ibm.com/quantum/blog/127-qubit-quantum-processor-eagle ↩
21. IBM Newsroom, "IBM Unveils 400 Qubit-Plus Quantum Processor and Next-Generation IBM Quantum System Two" (Nov 9, 2022), https://newsroom.ibm.com/2022-11-09-IBM-Unveils-400-Qubit-Plus-Quantum-Processor; and Madsen, L. S., et al., "Quantum computational advantage with a programmable photonic processor" (Xanadu Borealis), Nature 606, 75-81 (2022). ↩
22. IBM Newsroom, "IBM Debuts Next-Generation Quantum Processor & IBM Quantum System Two" (Dec 4, 2023). https://newsroom.ibm.com/2023-12-04-IBM-Quantum-Summit-2023 ↩
23. Atom Computing, "Atom Computing is the first to exceed 1,000 qubits" (Oct 24, 2023). https://atom-computing.com/quantum-startup-atom-computing-first-to-1000-qubits/ ↩
24. Quantinuum, "Quantinuum's H2 quantum computer achieves 99.914% two-qubit gate fidelity" (2024). https://www.quantinuum.com/ ↩
25. Microsoft Azure Quantum, "Microsoft unveils Majorana 1, the world's first quantum processor powered by topological qubits" (Feb 19, 2025), https://azure.microsoft.com/en-us/blog/quantum/2025/02/19/microsoft-unveils-majorana-1/; and Aghaee, M., et al. (Microsoft Quantum), "Interferometric single-shot parity measurement in InAs-Al hybrid devices," Nature 638, 651-655 (2025). ↩
26. Google Quantum AI, "Google Quantum AI shows verifiable quantum advantage with Quantum Echoes," Google Research blog and Nature (Oct 2025). https://blog.google/technology/google-deepmind/quantum-echoes-verifiable-quantum-advantage/ ↩
27. Nielsen, M. A., and Chuang, I. L., Quantum Computation and Quantum Information, Cambridge University Press (2010), chapters on entanglement and quantum algorithms. ↩
28. Krantz, P., et al., "A quantum engineer's guide to superconducting qubits," Applied Physics Reviews 6, 021318 (2019). https://doi.org/10.1063/1.5089550 ↩
29. Rigetti Computing, "Rigetti launches Ankaa-3 system, achieving 99.5% median two-qubit gate fidelity" (Dec 2024). https://investors.rigetti.com/news-releases ↩
30. Bruzewicz, C. D., Chiaverini, J., McConnell, R., and Sage, J. M., "Trapped-ion quantum computing: Progress and challenges," Applied Physics Reviews 6, 021314 (2019). https://doi.org/10.1063/1.5088164 ↩
31. IonQ, "IonQ Forte and the #AQ (algorithmic qubits) benchmark," IonQ. https://ionq.com/quantum-systems ↩
32. Quantinuum, "Quantinuum H2-1 achieves a quantum volume of 2^23 (8,388,608)" (2023). https://www.quantinuum.com/ ↩
33. Slussarenko, S., and Pryde, G. J., "Photonic quantum information processing: A concise review," Applied Physics Reviews 6, 041303 (2019). https://doi.org/10.1063/1.5115814 ↩
34. Henriet, L., et al., "Quantum computing with neutral atoms," Quantum 4, 327 (2020). https://doi.org/10.22331/q-2020-09-21-327 ↩
35. QuEra Computing, "Aquila: a 256-qubit neutral-atom quantum processor," QuEra. https://www.quera.com/aquila ↩
36. Fowler, A. G., Mariantoni, M., Martinis, J. M., and Cleland, A. N., "Surface codes: Towards practical large-scale quantum computation," Physical Review A 86, 032324 (2012). ↩
37. Cross, A. W., Bishop, L. S., Sheldon, S., Nation, P. D., and Gambetta, J. M., "Validating quantum computers using randomized model circuits," Physical Review A 100, 032328 (2019). https://doi.org/10.1103/PhysRevA.100.032328 ↩
38. Cao, Y., et al., "Quantum chemistry in the age of quantum computing," Chemical Reviews 119, 10856-10915 (2019). https://doi.org/10.1021/acs.chemrev.8b00803 ↩
39. Liu, M., et al., "Certified randomness using a trapped-ion quantum processor," Nature (2025), https://www.nature.com/articles/s41586-025-08737-1; JPMorganChase, "Advancing the application of quantum computing" (Mar 26, 2025). https://www.jpmorganchase.com/about/technology/news/certified-randomness ↩
40. Gidney, C., and Ekera, M., "How to factor 2048 bit RSA integers in 8 hours using 20 million noisy qubits," Quantum 5, 433 (2021). https://doi.org/10.22331/q-2021-04-15-433 ↩
41. Biamonte, J., et al., "Quantum machine learning," Nature 549, 195-202 (2017). https://www.nature.com/articles/nature23474 ↩
42. Preskill, J., "Quantum computing in the NISQ era and beyond," Quantum 2, 79 (2018). https://doi.org/10.22331/q-2018-08-06-79 ↩
43. McKinsey and Company, "Quantum Technology Monitor" (2025). https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/quantum-technology-monitor ↩
44. Gidney, C., "How to factor 2048 bit RSA integers with less than a million noisy qubits," arXiv:2505.15917 (May 2025). https://arxiv.org/abs/2505.15917 ↩
45. IBM Newsroom, "IBM Delivers New Quantum Processors, Software, and Algorithm Breakthroughs on Path to Advantage and Fault Tolerance" (Nov 12, 2025). https://newsroom.ibm.com/2025-11-12-ibm-delivers-new-quantum-processors,-software,-and-algorithm-breakthroughs ↩
46. Quantinuum, "Quantinuum announces commercial launch of new Helios quantum computer that offers unprecedented accuracy to enable Generative Quantum AI (GenQAI)" (Nov 5, 2025). https://www.quantinuum.com/press-releases/quantinuum-announces-commercial-launch-of-new-helios-quantum-computer ↩
47. IonQ, "IonQ achieves record-breaking quantum performance milestone of #AQ 64" (2025). https://investors.ionq.com/news/news-details/2025/IonQ-Achieves-Record-Breaking-Quantum-Performance-Milestone-of-AQ-64/ ↩
48. NVIDIA Newsroom, "NVIDIA introduces NVQLink, connecting quantum and GPU computing for 17 quantum builders and nine scientific labs" (Oct 28, 2025). https://nvidianews.nvidia.com/news/nvidia-nvqlink-quantum-gpu-computing ↩
49. PsiQuantum, "PsiQuantum announces Omega, a manufacturable chipset for photonic quantum computing" (Feb 26, 2025), and Alexander, K., et al., "A manufacturable platform for photonic quantum computing," Nature 641, 876-883 (2025). https://www.nature.com/articles/s41586-025-08820-7 ↩
50. Bausch, J., et al. (Google DeepMind and Google Quantum AI), "Learning high-accuracy error decoding for quantum processors" (AlphaQubit), Nature 635, 834-840 (2024). https://www.nature.com/articles/s41586-024-08148-8 ↩
51. Microsoft Azure Quantum, "Microsoft and Atom Computing offer a commercial quantum machine with the largest number of entangled logical qubits on record" (Nov 19, 2024). https://azure.microsoft.com/en-us/blog/quantum/2024/11/19/microsoft-and-atom-computing/ ↩
52. Hoefler, T., Haner, T., and Troyer, M., "Disentangling hype from practicality: On realistically achieving quantum advantage," Communications of the ACM 66(5), 82-87 (2023). https://doi.org/10.1145/3571725 ↩