Quantum computer Helios and IBM’s Loon: 2025 as the year of a quantum breakthrough

While the world is busy talking about ChatGPT, Sora, and AI avatars, another technological wave is quietly forming in the background – quantum computers. In November 2025, something happened that many researchers describe as a turning point: Quantinuum introduced the quantum computer Helios, currently the most powerful in the world, while IBM unveiled its experimental chip Loon, designed as a key step towards practical quantum machines by the end of this decade.

At the same time, Europe switched on its first two quantum processors, Jade and Ruby, directly integrated into existing supercomputers. All of a sudden, the quantum ecosystem has three big new players on the board – and it’s getting harder to say that quantum computing is “something for 2050.”

This article is for readers who follow AI, crypto and new technologies and are asking themselves: is a quantum computer still just a distant theory, or are we really approaching the moment when it will change how we build drugs, materials, and even AI algorithms?

Helios: the most powerful quantum computer on the planet

Quantinuum’s Helios is an ion-trap quantum computer – it uses trapped barium ions as qubits. What makes it stand out from earlier systems is a combination of three things:

• 98 physical qubits connected in a new “junction” architecture,
• the ability to form dozens of logical qubits with much lower error rates than previous generations,
• a brand-new software stack and programming model for quantum algorithms.

Instead of the usual ratio of 10:1 (around ten physical qubits needed to obtain one error-corrected logical qubit), Helios pushes that down to roughly 2:1. In other words, a much larger share of the quantum hardware is doing useful work, instead of being “sacrificed” purely for error correction.

In early tests, Helios has:

• set new records in standard quantum benchmarks,
• demonstrated very high fidelity of quantum operations,
• been used to simulate the behavior of a high-temperature superconductor and reveal quantum effects that are almost impossible to probe directly in the lab.

Unlike many “biggest ever” quantum devices from previous years, Helios doesn’t impress only with the number of qubits, but with quality and real results. That’s why a lot of researchers informally call it the first system that seriously hints at the coming fault-tolerant era – quantum machines that can run algorithms for hours or days without their errors spiraling out of control.

New software for new hardware

Helios doesn’t come alone – there’s a whole software layer built around it:

• a quantum programming language designed for this and future generations of hardware,
• a classical control system running on GPUs (Nvidia) that catches errors in real time and sends corrections back to the quantum chip,
• integration with existing simulation libraries for materials, chemistry and physics.

For developers that means we’ll increasingly see hybrid algorithms in the coming years – part of the computation runs on classical CPUs/GPUs, part on a quantum processor, and the two sides constantly exchange data.

IBM’s Loon: a roadmap to practical quantum machines by 2029

While Helios shows what’s possible right now, IBM is focused on a slightly different question: how do you go from an experimental system to a large-scale, practical quantum computer? One of their answers arrives in the form of a new chip called Loon.

Loon is an experimental processor that combines:

• regular (physical) qubits,
• additional quantum links between them,
• and a specialized architecture aimed at quantum error correction.

For years, IBM has been working on an approach where ideas from telecommunications – the same type of error-correcting codes that keep your mobile signal clean – are adapted for the quantum world. Loon is the first chip to show that this concept can actually be implemented in real hardware.

The plan looks roughly like this:

• Nighthawk – the next chip which should show quantum advantage over classical computers on specific tasks as early as 2026,
• continued development of architectures derived from Loon,
• the ultimate goal: a large, practical, fault-tolerant quantum computer around 2029.

One important detail is openness: IBM wants as many startups, universities and researchers as possible to experiment with their chips via a shared software ecosystem. Instead of chasing bombastic headlines, the focus is to find a realistic set of problems where a quantum computer truly helps – from financial optimization and logistics to AI modeling.

Europe joins the game: Jade and Ruby quantum processors

Quantum computing is not just a US-plus-private-companies story. The European Union is actively building its own ecosystem. The HPCQS project (High-Performance Computing and Quantum Simulator hybrid) switched on two quantum processors in November 2025:

• Jade in Germany,
• Ruby in France.

What makes them different from many other systems is how they’re integrated: both are directly connected to existing supercomputers in those centers. Instead of a quantum computer being an exotic machine in a separate building, Jade and Ruby act as accelerators inside familiar HPC environments.

This means European researchers in:

• climate modeling,
• simulations of materials and chemical reactions,
• optimization of energy grids,

can run hybrid workloads – part of the job on a classical supercomputer, part on a quantum processor – using the same tools they already know.

For industry, this sends a clear signal: Europe wants its own quantum stack – from hardware to software to applications – and doesn’t intend to rely entirely on American or Chinese companies.

What does a quantum breakthrough actually change?

1. AI and “quantum-enhanced intelligence”

Quantum computers don’t replace classical AI, but they can significantly boost it in certain tasks:

• training or optimization problems that require searching through gigantic parameter spaces,
• accelerated simulations of materials for AI-designed batteries, drugs, and chips,
• a new class of algorithms for cryptography and security.

For the average person, the difference won’t show up as “I own a quantum phone,” at least not any time soon. It will show up indirectly: faster drug discovery, smarter materials, more efficient energy networks.

Just like with the early internet – most people didn’t “see” the backbone infrastructure, they just saw email and websites. Quantum will likely follow a similar pattern.

2. Cryptography and the crypto-economy

The classic story “a quantum computer will instantly kill Bitcoin” is still far from reality – we’d need far more stable qubits than Helios or Loon currently have. But the direction of travel is clear:

• traditional crypto algorithms (RSA, ECC) will gradually need to move to post-quantum variants,
• blockchain projects that think long-term are already exploring migrations to quantum-resistant schemes,
• states and banks will have to move critical systems to new standards over the next decade.

For developers in the crypto space, this is a loud signal: keep an eye on post-quantum cryptography just as much as on new L2 networks.

3. Industry, science and everyday users

The biggest direct benefits will initially flow to:

• pharma (drug and protein design),
• chemistry and materials science (superconductors, batteries, lighter and stronger materials),
• finance and logistics (large-scale portfolio, route and supply-chain optimization).

For everyday users, that will look like:

• cheaper and more efficient devices (batteries, chips, materials),
• faster development of new treatments,
• smarter energy and transport networks.

As with the early AI wave, the most profound changes will happen “under the hood” – deep in the infrastructure and tools, not necessarily as a gadget on your desk.

Conclusion

We’ve been hearing for years that quantum computers are “always ten years away.” What 2025 brings is a very different feeling: that distance is finally shrinking.

• Helios shows that serious error correction and real scientific applications are already possible today,
• IBM’s Loon and the roadmap to fault-tolerant machines by 2029 give industry a concrete time horizon,
• Jade and Ruby put Europe firmly on the map with hybrid supercomputers.

In the next five to ten years we probably won’t have a quantum laptop at home, but it’s entirely realistic that:

• quantum modules will become a standard part of supercomputers,
• AI, pharma, energy and finance will gain completely new tools,
• post-quantum cryptography will become a mandatory piece of digital infrastructure.

For InfoHelm Tech readers, the takeaway is simple: if you’re interested in a career at the intersection of physics, math, programming and AI, quantum computing is rapidly moving from science fiction into a very real tech niche. And we’ll keep following every step along that path.

Disclaimer: This article is for informational purposes only and does not constitute financial, investment, legal, career, or any other kind of professional advice.