Quantinuum’s new ion-based quantum computer, Helios, marks a significant step forward in the pursuit of scalable and reliable quantum computing. It could change how we handle error correction in the quantum age.
A Major Step in Quantum Hardware
US- and UK-based company Quantinuum has introduced Helios, its third-generation quantum computer. This represents a major advance in the company’s plan for scalable and commercially viable quantum systems. Unlike Google’s and IBM’s quantum computers, which use superconducting circuits, Helios employs trapped ions, or electrically charged atoms, as its qubits. While this hardware approach is complex, it offers a major benefit: improved accuracy and easier error correction.
Helios is an important proof point in our roadmap about how we’ll scale to larger physical systems,” says Jennifer Strabley, Vice President at Quantinuum.
Founded in 2021 through the merger of Honeywell Quantum Solutions and Cambridge Quantum, Quantinuum is largely owned by Honeywell. With Helios, the company believes that trapped-ion technology could surpass other architectures in both stability and scalability.
Inside Helios: 98 Barium Ions at Work
The new Helios system, located at Quantinuum’s facility in Colorado, uses precise optics, lasers, and mirrors in a tightly controlled environment kept at around 15 Kelvin (-432.67°F).
At its center, there is a thumbnail-sized chip that contains 98 barium ions, which serve as the computer’s qubits. These ions are suspended and manipulated with electromagnetic fields and laser light to perform quantum operations with high stability.
This represents a significant upgrade from Quantinuum’s previous system, H2, which used 56 ytterbium ions. Switching to barium ions enhances the system’s control fidelity, allowing for more accurate operations with fewer mistakes. Users can access Helios remotely via the cloud, letting researchers and organizations run quantum algorithms from anywhere.
Quantum States: Beyond Classical Bits
In traditional computing, information is recorded in bits that can be either 0 or 1. In quantum computing, however, qubits can exist in a combination of both states at the same time, a phenomenon called superposition.
Imagine flipping a coin. When it is in the air, it is neither just heads nor just tails; it is a blend of both. This is similar to a qubit’s state. Quantum computers use these unique properties to perform calculations that would take classical computers significantly longer.
The barium ions in Helios store information in their quantum states. Through quantum entanglement, these ions can be connected so that the state of one qubit affects the state of another, even if they are far apart. This entanglement is vital for quantum computation, allowing for massively parallel data processing.
The Race Toward Useful Quantum Advantage
Despite advancements in quantum hardware, no quantum computer today can perform commercially valuable tasks beyond what classical machines handle. Still, companies like Quantinuum, Google, IBM, and IonQ are making progress.
Helios represents one of the most advanced ion-trap systems created so far. It serves as a platform for scaling quantum systems and improving error correction, a key obstacle that hinders quantum computers from reaching their commercial potential.
Researchers and industry experts see quantum computers changing fields such as materials discovery, battery chemistry, drug design, and financial modeling. However, achieving this vision depends on overcoming the notoriously high error rates in quantum systems.
The Quantum Error Problem
All computers make mistakes, but in quantum computing, errors pose a much larger issue. Quantum systems are highly sensitive to their environment; even minor disturbances can cause decoherence, disrupting a qubit’s delicate quantum state.
Classical computers manage errors through redundancy by copying information across multiple bits. Quantum computers cannot directly copy qubit data (due to the no-cloning theorem), so they must use more complex quantum error correction (QEC) methods.
QEC works by spreading a single “logical qubit” across multiple physical qubits, allowing the system to identify and fix errors without measuring or destroying the fundamental quantum state.
Helios’ Simpler Approach to Error Correction
Helios stands out because it only requires two physical qubits to create one logical qubit, which is much less overhead than other architectures need.
For comparison:
– Google (2024) needed 105 superconducting qubits to create one logical qubit.
– IBM (2025) accomplished this with 12 physical qubits.
– Amazon Web Services used nine physical qubits.
Helios’ two-to-one ratio is significantly more efficient, indicating a substantial breakthrough in error-corrected quantum computation. This efficiency originates from the natural stability of trapped ions, which generally exhibit lower error rates than superconducting circuits.
Quantinuum reports that pairs of its qubits, when entangled, function correctly 99.921% of the time — one of the highest accuracy rates ever recorded in a quantum processor.
“To the best of my knowledge, no other platform is at this level,” says Rajibul Islam, a physicist at the University of Waterloo, who is not affiliated with Quantinuum.
All-to-All Connectivity: A Hidden Advantage
Another key feature of Helios is its “all-to-all connectivity.” In superconducting systems, qubits are fixed on a chip and can only interact with nearby qubits. Connecting distant qubits requires extra intermediate operations, each adding potential errors.
In contrast, trapped ions can move. On the Helios chip, ions can be physically rearranged, allowing any qubit to interact directly with any other qubit.
This flexibility simplifies quantum circuit design and reduces the number of operations needed for complex calculations, making the process more reliable.
“It’s becoming increasingly apparent how important all-to-all connectivity is for these high-performing systems,” notes Strabley.
Competing Architectures: The Quantum Diversity Problem
While Helios shows impressive performance, the quantum computing landscape remains fragmented. Each hardware type has its own strengths and weaknesses, and it is still unclear which architecture will prevail.
– Trapped ions (Quantinuum, IonQ): high fidelity, but complex and delicate to operate.
– Superconducting circuits (IBM, Google, AWS): easier to manufacture, but more error-prone.
– Neutral atoms (QuEra): simpler to trap and possibly more scalable, but still experimental.
As Islam states, “Even with fewer physical qubits, ions let you do more.” However, manufacturing ion-based systems is slow and costly, whereas superconducting qubits can be mass-produced using existing semiconductor methods.
Error Correction on the Fly
One of the most innovative aspects of Helios is its ability to perform “error correction on the fly.”
Traditional QEC requires constant post-processing and monitoring, but Helios incorporates real-time error detection powered by Nvidia GPUs.
According to David Hayes, Quantinuum’s Director of Computational Theory and Design, GPUs enable parallel error detection, making the process faster and more scalable than the FPGAs used in most competing systems.
This capability is a vital step toward autonomous, self-correcting quantum processors, a key milestone in achieving fully reliable computing.
Early Experiments: Magnetism and Superconductivity
Beyond hardware, Quantinuum is already using Helios for scientific research.
The company previously employed its H2 system to simulate magnetic phenomena, achieving results that rival the best classical methods.
With Helios, researchers have advanced further, simulating how electrons behave in high-temperature superconductors. This fundamental physics problem has significant implications for energy and materials science.
“These aren’t contrived problems,” says Hayes. “These are problems that the Department of Energy, for instance, is very interested in.”
Such simulations show that even before reaching commercial-scale quantum advantage, quantum machines are already enhancing our understanding of nature.
Building the Future: Sol and Apollo
Quantinuum is not stopping with Helios. The company has announced plans for its next-generation systems, continuing its mythological naming theme.
– Sol (2027): Expected to have 192 physical qubits, doubling Helios’s capacity.
– Apollo (2029): Aiming for thousands of qubits and full fault tolerance, which would allow it to detect and correct all errors in real time.
A second Helios machine is already being built at Quantinuum’s facility in Minnesota, ensuring redundancy and scalability for future research and enterprise users.
Why Helios Matters
Helios does not yet achieve the commercial quantum breakthroughs that investors desire. However, it may represent something more significant: a credible path toward scalable and reliable quantum computing.
Its trapped-ion design, low error-correction demands, and real-time fault management make it one of the leading platforms in the field.
If Quantinuum can keep up its pace, the Helios-Sol-Apollo roadmap could redefine what is possible in practical quantum computing by the end of the decade.
For now, Helios stands as both a scientific achievement and a technological claim: error correction does not need to block quantum computing; it can be a bridge.
