Quantum computing’s six most important trends for 2025

Quantum computing’s six most important trends for 2025

The financial industry is anticipated to become one of the earliest adopters of commercially useful quantum computing technologies. These technologies are expected to become available within the next few years, making it more important than ever to follow experimental developments. Therefore, as 2024 comes to a close, we pause to reflect upon what has recently transpired and look for trends we believe will continue through 2025 and beyond. This exercise has led to the identification of six trends, each of which we cover through 2024 reports and announcements:

1. More experiments with logical qubits
   
   
2. More specialized hardware/software (as opposed to universal quantum computing)
   
   
3. More networking noisy intermediate-scale quantum (NISQ) devices together
   
   
4. More layers of software abstraction
   
   
5. More workforce development tools
   
   
6. Improved and novel physical qubits
    

These trends are not ordered by significance or likelihood; we see all of these as equally likely.
 

More experiments with logical qubits

In “The road to useful quantum computing,” published on February 13, 2024, we reviewed the events of 2023 that started shifting our conversations from physical qubits in 2022 to logical qubits in 2024. Whereas a “physical qubit” refers to actual error-prone hardware, a “logical qubit” is an arrangement of these physical qubits that encodes information in such a way as to protect against errors [1]. We chronologically cited 19 achievements — many of which were experimental — from 23 organizations; alphabetically, they are: Alice & Bob, Amazon Web Services, Caltech, Chinese Academy of Sciences, Cornell, ETH Zurich, Fuzhou University, Google, Harvard, IBM, MIT, PsiQuantum, QC Design, Quantinuum, QuEra Computing, QuTech, Riverlane, Tsinghua University, the University of Chicago, the University of Science and Technology of China, the University of Sheffield, the University of Stuttgart,  and Yale. We noted that by early 2024, QuEra and Infleqtion had already published their logical qubit roadmaps, while Alice & Bob had teased that a logical qubit roadmap would be forthcoming [2]. 

In 2024, we observed continued experimentation:

• On August 27, Google announced that it had demonstrated a quantum memory with below-threshold error rates and double the coherence lifetimes as compared with physical qubits [3]. 
  
  
• On September 10, Microsoft and Quantinuum announced that they had entangled 12 logical qubits, triple the logical qubit count from six months prior [4]. The physical error rate of 0.024 was brought down to a logical error rate of 0.0011, leading to the first ever chemistry simulation combining high-performance computing (HPC), artificial intelligence (AI), and quality control (QC) [5]. 
  
  
• On October 31, Rigetti and Riverlane announced that they had demonstrated error correction with gate speeds fast enough to support heterogeneous quantum-classical processing [6].  
  
  
• On December 3, an IBM Quantum team announced that they had demonstrated the entanglement of logical qubits using overlapping codes, applying a method thought to be impractical with limited connectivity [7]. 
  
  
• On December 9, Google announced that its Willow chip demonstrated below-threshold error correction, lowering error rates as more physical qubits encode logical qubits [8]. 
   

We’ve also seen growth in the number of logical qubit roadmaps. In addition to the three companies we mentioned, Inside Quantum Technology mentioned IBM Quantum, Google, and Microsoft in “The quantum roadmap battle of logical qubits,” which was published on February 27, 2024 [9]. Since that time, we have added eight more to this list (alphabetically, given that not all of them are dated):

• Diraq [10] 
  
  
• IonQ [11] 
  
  
• IQM [12] 
  
  
• Pasqal [13] 
  
  
• PsiQuantum [14] 
  
  
• Quantinuum [15] 
  
  
• Quandela [16] 
  
  
• Xanadu [17] 
   

While this may not be an exhaustive list, it shows a continuing trend either way. We’ve gone from talking about physical qubits to demonstrating logical qubits to showing that logical qubits lower error rates. The growth in logical qubit roadmaps requires accelerated experimentation, especially considering that quantum computers based on logical qubits are being promised within the next few years. 
 

More specialized hardware/software (as opposed to universal QC)

Building a universal quantum computer — a quantum computer that can run any quantum algorithm — is hard [18]. If it were easy, we would already have one. While this is still the end goal, multiple companies are developing specialized quantum computers for specific problems to achieve earlier commercial value:

• Bleximo is building full-stack superconducting application-specific systems with co-designed processors, software, and control stacks [19]. 
  
  
• Qilimanjaro is building quantum app-specific integrated circuits (QASICs) for superconducting analog quantum computers with full stacks co-designed for specific problems [20]. 
  
  
• QuiX is building special-purpose photonic quantum computers for specific optimization and simulation problems [21]. 
  
  
• On November 19, QuEra launched a full-stack quantum algorithm co-design program to optimize hardware, software, and applications for specific problems [22]. 
   

It’s reasonable to assume that this list will continue to grow. Likewise, it’s presumably easier to sell a special-purpose quantum computer that provides a commercial advantage for one or a few problems than it is to sell a general-purpose quantum computer that does not yet provide commercial advantages for any problems.
 

More networking NISQ devices together

Two ways to scale up a quantum computer are to add more qubits to a single quantum computer or to interconnect multiple quantum computers, creating a single virtual quantum computer with a higher qubit count. In 2024, several companies announced larger individual processors, but there were also developments in connecting multiple processors:

• On May 30, Photonic announced that it had demonstrated distributed entanglement, linking qubits within separate quantum computers [23]. 
  
  
• On October 30, QuTech announced that it had connected two small quantum computers in two different cities [24]. 
  
  
• IBM is building L-couplers to scale its systems [25]. On November 21, it announced that it had classically linked two 127-qubit quantum processors to create a virtual 142-qubit system [26]. 
   

There is far more research going into quantum networks than the examples here, but not all of it links to quantum computers just yet. Researchers are still developing the infrastructure. However, experimental demonstrations of interconnected and internetworked quantum computers are bound to lead to experimentation with distributed applications using more of these nascent quantum networks.
 

More layers of software abstraction

A major challenge in adopting quantum computers is learning how to use them. Quantum algorithms and quantum hardware are significantly different from their classical counterparts. Fortunately, multiple companies are designing interfaces so users might not need to know anything about quantum computers: 

• Multiverse Computing’s Singularity uses a spreadsheet to abstract away whether it is using quantum or quantum-inspired computing to solve optimization problems [27]. 
  
  
• The Strangeworks model allows you to define your problem. Once you’ve done that, you simply select a solver — quantum or otherwise — and run it [28]. 
   

Quantastica’s Quantum Algorithm Generator doesn’t completely abstract quantum technology, but it converts a classical function into a quantum circuit [29]. This lowers the barrier to entry but still requires some knowledge of quantum computing.

Other companies are attempting to be hardware-agnostic, removing the hardware layers so it’s easier to develop cross-platform applications. However, this still requires a considerable amount of quantum computing knowledge. These solutions are halfway there, though, and simply need the software component to join this list.
 

More workforce development tools

There is no shortage of introductory quantum computing materials, but the vast majority are self-paced or structured for individuals. At least two companies offer programs with enterprises in mind, recognizing the need for workforce development rather than just career development:

• Q-CTRL’s Black Opal is an interactive online learning platform that individuals can use, but it can also be embedded into a formal curriculum with a capstone project, exams, and automatic grading [30]. 
  
  
• QURECA offers customized training programs tailored to organizational needs, various roles, and different business sectors [31]. There is also a team-building program that bridges quantum computing hard skills and enterprise soft skills [32]. 
   

MIT offers a two-course program from MIT xPRO that is designed for working professionals and awards a certificate. Professor Peter Shor, the developer behind Shor’s algorithm, and Professor Isaac Chuang, co-author of “The Bible of quantum computing,” are two of the instructors [33]. It is easy to imagine similar programs being adapted to individual enterprises, where an entire course consists of employees from the same organization. 
 

Improved and novel physical qubits

Two ways to lower error rates on quantum computers are to improve the physical qubits and to implement high-overhead error correction codes. It is widely believed that quantum error correction (QEC) will be necessary; however, research continues into improving physical qubits, which in turn boosts the performance of the logical qubits: [34] 

• On May 6, the University of Basel announced the demonstration of hole spin qubits, which use the spins of missing electrons in semiconductors and require fewer components than electron spin qubits [35]. 
  
  
• On September 19, University College Dublin announced that "split-electrons," so-called “Majorana fermions,” can theoretically be used as topological qubits [36]. 
  
  
• On September 21, Brookhaven National Laboratory announced it had used constriction junctions, which simplify chip fabrication to enable mass production [37]. 
  
  
• On November 26, The Quantum Insider reported that Quantinuum, Harvard, and Caltech experimentally demonstrated the first “true topological qubit,” robustly encoding information in systems’ relationship patterns instead of within the systems [38]. 
  
  
• On November 26, as part of the Q-LEAP project, RIKEN and Toshiba announced a double-transmon coupler (DTC), which has been proposed to improve gate fidelity [39]. 
  
  
• On November 26, The Wall Street Journal reported that Ephos uses glass chips instead of silicon to reduce energy consumption and information loss [40]. 
  
  
• On December 4, the American Physical Society reported in its Physics publication that ETH Zurich demonstrated a mechanical qubit consisting of two sapphire chips, with a superconducting qubit on top with a mechanical resonator below [41]. 
   

It’s impossible to know everything that is being researched at any given time. We can assume, however, that research continues in universities and labs around the world, as well as at many quantum computing companies.
 

Conclusion

While no one has a crystal ball, we can confidently predict 2025 trends based on analyses of 2024 and 2023. The exciting part of this is the unknowns. If we were in 2022, we probably wouldn’t be predicting the experimental demonstration of logical qubits in 2023, but that happened. We’re excited to find out what we’ve missed, and we hope you’ll join us throughout the year as we continue along the path toward fault-tolerant quantum computing.
 

References

1. QuEra Computing, “Logical Qubit,” no date.
2. Moody’s, “The road to useful quantum computing,” February 13, 2024.
3. Swayne, Matt, “Breaking The Surface: Google Demonstrates Error Correction Below Surface Code Threshold,” The Quantum Insider, August 27, 2024.
4. Zander, Jason, “Microsoft announces the best performing logical qubits on record and will provide priority access to reliable quantum hardware in Azure Quantum,” The Official Microsoft Blog, September 10, 2024.
5. Svore, Krysta, “Microsoft and Quantinuum create 12 logical qubits and demonstrate a hybrid, end-to-end chemistry simulation,” Microsoft Azure Quantum Blog, September 10, 2024.
6. Choucair, Cierra, “Rigetti and Riverlane achieve real-time quantum error correction on 84-qubit system,” The Quantum Insider, October 31, 2024.
7. Hetényi, Bence; Wootton, James, “Creating entangled logical qubits in the Heavy-Hex lattice with topological codes,”  PRX Quantum, December 3, 2024.
8. Neven, Hartmut, “Meet Willow, our state-of-the-art quantum chip,” The Keyword, December 9, 2024.
9. Siegelwax, Brian, “The quantum roadmap battle of logical qubits by Brian Siegelwax,” Inside Quantum Technology News, February 27, 2024.
10. Diraq, “Roadmap,” no date.
11. IonQ, “IonQ unveils accelerated roadmap and new technical milestones to propel commercial quantum advantage forward,” July 22, 2024.
12. IQM Finland Oy, “IQM quantum computers unveils development roadmap focused on fault-tolerant quantum computing by 2030,” November 13, 2024.
13. Pasqal, “Our roadmap,” no date.
14. PsiQuantum, “Blueprint,” no date.
15. Quantinuum, “Quantinuum unveils accelerated roadmap to achieve universal, fully fault-tolerant quantum computing by 2030,” September 10, 2024.
16. Quandela, “Leading the way towards the fault-tolerant era,” no date.
17. Xanadu, “Millions of qubits powered by light,” no date.
18. Wikipedia, “Quantum Turing Machine," December 5, 2024.
19. LinkedIn, Bleximo Corp,  no date.
20. Qilimanjaro Quantum Tech, “Home,” no date.
21. QuiX Quantum, “Special purpose photonic quantum computer: Quix Quantum,” [M(1] no date. 
22. QuEra Computing, “QuEra launches full-stack quantum algorithm co-design program to maximize quantum computing potential,”[M(2]  November 19, 2024.[M(3] [A(4] 
23. Photonic, “Photonic demonstrates distributed entanglement between modules, marking significant milestone toward scalable quantum computing and networking,” May 30, 2024.
24. QuTech, “A rudimentary quantum network link between Dutch cities,” October 30, 2024.
25. Freund, Karl, “IBM launches Quantum System Two and a roadmap to quantum advantage,”[M(5] [M(6]  Forbes, December 4, 2023.
26. Swayne, Matt, “The classical-quantum connection: Scientists link quantum processors with real-time classical connection,” The Quantum Insider, November 21, 2024.
27. Multiverse Computing, “Singularity,”[M(7] [M(8]  no date.
28. Strangeworks, “Strangeworks optimization,” no date.
29. Quantastica, “Quantum algorithm generator,”no date.
30. Q-CTRL, “Learn quantum computing,”[M(9] [M(10]  no date.
31. Qureca, “Quantum training for business,”[M(11] [M(12]  no date.
32. Qureca, “Quantum Team Building Experience,” [M(13] no date.
33. MIT xPro, “Quantum Computing Fundamentals,” [M(14] Massachusetts Institute of Technology, no date.
34. Neven, Hartmut; Kelly, Julian, “Suppressing quantum errors by scaling a surface code logical qubit,” Google Research, February 22, 2023.
35. Caluori, Reto, “Experiment opens door for millions of qubits on one chip,” Phys.org, May 6, 2024.
36. University College Dublin, “Topological quantum computers a step closer with new method to ‘split’ electrons,” Phys.org, September 19, 2024.
37. Brookhaven National Laboratory, “Brookhaven Lab’s new qubit architecture could change quantum computing forever,” SciTechDaily, September 21, 2024.
38. Swayne, Matt, “Research team achieves first-ever topological qubit, a step along the path toward fault-tolerant quantum computing,” The Quantum Insider, November 26, 2024.
39. RIKEN, “Scientists develop novel high-fidelity quantum computing gate,” November 26, 2024.
40. Forbes, Don Nico, “Glass chips offer hope of cleaner future for quantum computing,”[M(15]  The Wall Street Journal, November 26, 2024.
41. Curtis, Susan,  “Enter the mechanical qubit,” Physics, [M(16] December 4, 2024.