Quantinuum

Overview

Quantinuum is one of the largest dedicated quantum computing companies in the world. It was formed in 2021 when Honeywell combined its quantum hardware business with Cambridge Quantum, a British software and algorithms company, and it now sells access to trapped-ion computers, quantum chemistry software, and a certified quantum random number generator. Its machines are used by pharmaceutical, chemical, financial, and government customers, as well as by research teams at national laboratories.

The company is best known for quality rather than raw qubit count. Its systems have repeatedly set records for how accurately two qubits can be entangled, and its researchers regularly publish error-correction experiments that other companies later try to match. That reputation helped it complete an initial public offering on Nasdaq in June 2026 under the ticker QNT.

History and background

The hardware side traces back to Honeywell, which spent years developing a trapped-ion architecture called the quantum charge-coupled device (QCCD) and launched its first commercial system in 2020. Cambridge Quantum, founded by Ilyas Khan, built the TKET compiler, the InQuanto chemistry platform, and the Quantum Origin randomness product. In November 2021 the two businesses merged into Quantinuum, with Honeywell holding a majority stake.

Since the merger the company has released a series of machines, from the 20-qubit H1 and the 56-qubit H2 to the Helios system launched in late 2025. It also worked with Microsoft on error-corrected logical-qubit demonstrations in 2024. In 2026 the business reorganized under a new holding company, Quantinuum Inc., and priced its IPO on June 3, 2026 at 60 dollars per share, selling 28 million shares in an upsized offering before trading began on June 4.

How the technology works

A trapped-ion qubit is a single charged atom suspended in a vacuum by electric fields and manipulated with lasers. Every atom of a given element is identical, so ion qubits are uniform and naturally long-lived, and gates between them can be very precise. Quantinuum uses a QCCD design in which ions are physically shuttled between different zones of a chip: some zones store qubits, others perform gates, and others read out results. Because ions can be moved next to any partner, the machine achieves flexible connectivity without long chains of swap operations.

The trade-offs are speed and engineering complexity. Moving ions takes time, so Quantinuum's circuits run more slowly than those on superconducting chips, and building the laser delivery, trap fabrication, and control systems is difficult. The company argues that the quality of each operation matters more than speed because error correction is limited by fidelity. In 2025 it reported a two-qubit gate fidelity of 99.921% on Helios, and in its second-quarter 2026 update it said it had demonstrated logical-qubit fidelity close to five nines, meaning errors on the order of one in a hundred thousand.

Quantinuum's published roadmap runs through three generations. Helios arrived in 2025, a larger system called Sol is planned for 2027, and a universal fault-tolerant machine called Apollo is targeted for 2029. These are company goals, and the later machines depend on solving scaling problems that have not yet been shown at full size.

Products and platforms

Processors in our Hardware Database

• Quantinuum H2 — 56 qubits, released 2023 (32 qubits); upgraded to 56 qubits in 2024

Key milestones

Strengths and challenges

Strengths

• Among the highest published gate and logical-qubit fidelities of any platform.
• A flexible QCCD architecture with strong connectivity between qubits.
• A software business, including TKET and InQuanto, that is independent of its own hardware.
• Over 2 billion dollars in cash after the IPO, giving it room to invest through the next hardware generations.

Challenges and open questions

• Trapped-ion operations are slower than superconducting ones, which matters for deep algorithms.
• Scaling beyond a hundred or so qubits requires more ion shuttling, lasers, and control hardware.
• The company is still far from profitable, and its valuation depends on roadmap targets (Sol in 2027, Apollo in 2029) that have not yet been met.
• In the market data we reviewed, the stock was trading below its IPO price, which shows how volatile expectations around quantum companies can be.

Why it matters

Quantinuum's results are among the best evidence for how error-corrected quantum computing might work in practice. When researchers talk about logical qubits that outperform their physical components, a large share of the demonstrations come from Quantinuum hardware, either alone or with partners such as Microsoft.

Its IPO also matters as a market signal. It was one of the biggest listings in the sector, and its results will be watched as a test of whether investors will fund long-term fault-tolerant quantum computing at scale. Readers comparing technologies should read this page alongside IonQ and IBM: trapped ions and superconducting circuits are the two leading approaches, and Quantinuum represents the quality-first end of the trapped-ion camp.

Latest news

• Quantinuum reports Q2 2026 revenue up 279% year over year and raises its full-year outlook — August 2026
• Quantinuum announces a partnership with Oracle to deliver Helios as a service on Oracle Cloud Infrastructure — 2026
• Quantinuum shows near five-nines logical fidelity on Helios — 2026
• Quantinuum completes its upsized Nasdaq IPO at 60 dollars per share — June 2026

Frequently asked questions

This profile is written for educational purposes and summarizes public announcements, company filings, and published research. Last reviewed: September 30, 2026. It is not investment advice. Check the company's official channels for the latest figures and claims.

← Back to all quantum companies