Quantum computing has long promised to transform industries by solving complex problems beyond the reach of classical computers. But with the sector rapidly evolving, we are now starting to appreciate the potential for real-world applications.
Finnish startup QMill is targeting a critical gap by addressing the limitations of current quantum computing hardware.
Founded in June 2024. The team is led by seasoned experts in quantum physics, mathematics, and computer science, including three Professors, six PhDs and one EMBA across the fields of maths, physics, computer science, and business.
The company has already secured a €4 million seed round and a €1 million grant and today announced that by Peter Sarlin, co-founder and SVP of AMD Silo AI has joined them an angel investor.
I spoke to three of its co-founders: CEO Dr Hannu Kauppinen, CTO Dr Ville Kotovirta and Chief Scientist Prof Mikko Möttönen.
Bringing commercial-ready algorithms to Noisy Intermediate-Scale Quantum (NISQ) computers
QMill's approach is to develop commercially viable quantum algorithms for Noisy Intermediate-Scale Quantum (NISQ) computers, the class of quantum devices expected to be available in the near term.
Noisy Intermediate-Scale Quantum (NISQ) computers typically feature tens to a few hundred qubits, and significant levels of quantum noise and error. These devices are not yet capable of continuous quantum error correction, which is necessary for fault-tolerant computing.
QMill's patented algorithms allow them to leverage near-future hardware to overcome the limitations of NISQ computers rather than wait for more advanced, error-corrected quantum computers.
According to Kotovirta:
"For the first time now, we are revealing that we have algorithms that promise quantum advantage, that can work faster than supercomputers."
Tackling verification challenges in cloud quantum computing
One of the biggest concerns with current quantum computing services is verification—ensuring that the calculations performed by a quantum computer are genuinely quantum in nature.
According to Kauppinen:
"If you access quantum computers, you probably access them through the cloud because people and organisations rarely have their own quantum computers. People pay hundreds of millions of euros every year to access these quantum computers.
And in principle, they don't have the security that they are real quantum computers at the moment.
You want to know whether you are really paying for using a real quantum computer or whether somebody's just pretending to be a quantum computer because it's much more costly to use a quantum computer than a classical computer."
Many early quantum systems have high error rates, with one in a thousand operations failing. QMill's newly developed verification algorithm, which is currently patent pending, offers an unbroken solution for verifying early-stage quantum computations.
QMill's new quantum algorithms have achieved impressive results.
Using the algorithm, a 200-qubit quantum computer with 99.99 per cent accuracy can find the solution in a second. Such a machine is expected to be commercially available in 2029.
In contrast, solving the same problem with the best conventional logic algorithm on a supercomputer takes more than ten minutes.
If the number of qubits is increased to 290, it takes the supercomputer over a year to solve the problem, although the quantum computer can solve it in an hour.
In other words:
• For n = 200 qubits, the QC takes 1 second, and Lumi takes 11.7 minutes
• For n = 246 qubits, the QC takes 1 minute, and Lumi takes 3.2 days
• For n = 287 qubits, the QC takes 1 hour, and Lumi takes 566.7 days
Why now for quantum computing?
Quantum computing is reaching an inflection point.
Kauppinen asserts that "the growth is mainly due to public investments. National research centers and universities buy quantum computers or buy time using quantum computers to develop their algorithms."
For example, this week, the VTT Technical Research Centre of Finland and IQM Quantum Computers completed and launched Europe's first 50-qubit superconducting quantum computer. It's now open to researchers and companies through the VTT QX quantum computing service.
Kauppinen shared:
"We see that there's an inflection point when we actually achieve quantum advantage—when powerful enough quantum computers meet the right kind of algorithms."
According to Möttönen, quantum computing is at a sweet spot for startups:
"If we had done this five years ago, the future of quantum computing would have looked too distant for investors. If we waited five more years, we wouldn't be the first—we would lose the first-mover advantage.
The sweet spot was exactly now—just a couple of years before commercial quantum advantage emerges."
Real-world applications
The company targets a variety of sectors. For telecom networks, QMill focuses on network planning, optimisation, and data analytics to minimise energy and site costs while maximising capacity and load efficiency.
Its algorithms can help with capacity planning, energy production, storage, and pricing optimisation in energy networks. For logistics and distribution, use cases include optimising route planning, delivery modes, and capacity planning to reduce distribution time and costs.
Additionally, QMill's algorithms can enhance real-time situational awareness in security and defence by analysing large-scale sensor data for optimal outcomes. Its algorithms may also serve as efficient subroutines in materials development, for which the company is open to partnerships.
QMill's software lowers barriers to entry
Rather than building quantum hardware, QMill is focusing on a software-driven business model. Its revenue model revolves around selling access to their proprietary quantum algorithms, by licensing the algorithm library for direct integration into customer systems and offering API-ased access to run algorithms on NISQ computing platforms.
This approach enables businesses to leverage quantum computing without needing deep in-house expertise.
With recent funding, a skilled team, and its patented (and patent-pending) algorithms, QMill is bringing quantum computing to critical industries for real-world uses in the near future.
Lead image: QMill. Photo: uncredited.
Would you like to write the first comment?
Login to post comments