Looking ahead: Quantum computing may finally be crossing from the lab into the real world. Quantinuum, one of the world's most valuable quantum startups, has unveiled its latest system called Helios, a machine the company claims delivers a new benchmark in reliability and scale. This launch means to signal that quantum is edging closer to solving practical problems for industries from finance to materials science – and to proving that this field's decades of promise might soon pay off.
The Helios hardware houses 98 physical qubits. But practical quantum computing isn't measured merely in quantity: physical qubits are vulnerable to minute environmental disturbances, which can produce errors even during short computations. The industry instead often focuses on "logical" qubits, which result when many physical qubits are interlinked via error-correction protocols to create stable, reliable quantum states.
What sets Helios apart, according to Quantinuum executives and outside quantum experts, is its ability to deliver 48 logical error-corrected qubits from 98 physical ones. Achieving a direct conversion with a nearly 2:1 ratio is notable, given that rival architectures require tens or hundreds of physical qubits to yield a single logical qubit. Producing more logical qubits from fewer physical ones is technically demanding, as it entails complex trade-offs across software design, hardware stability, and error-correction algorithms.
"That's why we call it Helios, because Helios is the sun, and we said this is dawning the commercial adoption phase of quantum computing in industry," Quantinuum President and Chief Executive Rajeeb Hazra told The Wall Street Journal.
Quantinuum emerged in 2021 after the merger of Cambridge Quantum Computing and Honeywell's quantum unit, combining assets drawn from advanced research and high-performance industrial engineering. The company's funding earlier this year – a $600 million equity round at a $10 billion valuation – confirmed its position at the top of the private quantum sector as competitors race to build larger, more commercially ready machines.
Quantinuum's roadmap projects another milestone by 2029: the Apollo quantum system, planned to feature thousands of physical qubits and hundreds of logical qubits. However, the company asserts that Helios already represents a substantial step toward fault-tolerant quantum computing – supporting reliable results for enterprise-scale problems that classical computers have long considered unsolvable.
A key technical enhancement in Helios is its new programming language, Guppy, designed to let engineers craft and optimize quantum algorithms directly for its architecture. Unlike previous quantum software tools, which were sometimes limited by hardware constraints and scaling challenges, Guppy is purpose-built to help developers transition their research from small-scale current systems to much larger machines as they come online.
Quantinuum's leadership contends that algorithms running successfully on Helios today will have a direct path to scale up when machines like Apollo are deployed, forming a modular approach for quantum software development.
Financial services firms such as JPMorgan Chase are among the early adopters of Helios during its private preview phase. The bank's global technology applied research group has already executed more advanced quantum algorithms on the platform, including those focused on optimizing real-time analytics for massive datasets. JPMorgan and its peers say they continue to collaborate with multiple quantum vendors – validating, benchmarking, and refining models as hardware capabilities expand.
Rob Otter, JPMorgan's head of global applied technology research, notes that the rapid advancement of underlying quantum systems could deliver enterprise-ready quantum workflows sooner than industry timelines had predicted. While large-scale fault-tolerant quantum computing remains several years away, firms are now laying the algorithmic foundations and infrastructure to capitalize as soon as quantum hardware matures.