Skip to main content
Azure
5 min read

Microsoft and Atom Computing offer a commercial quantum machine with the largest number of entangled logical qubits on record

Microsoft and Atom Computing have made rapid progress in reliable quantum computing by creating and entangling 24 logical qubits made from neutral atoms—they have also demonstrated the ability to detect and correct errors, and perform computation, on 28 logical qubits. Together, the companies are offering a reliable quantum machine built with these state-of-the-art logical qubits and Microsoft’s qubit-virtualization system, and integrated with Azure Elements. This comprehensive package is available to order today with delivery in 2025. The results reported here showcase the advances being made toward scientific quantum advantage, which will require not only reliable quantum computation, but also integration with AI and high-performance computing.

Atom Computing’s neutral-atom qubits

In the quantum computing industry, multiple types of physical qubits are being explored as potential pathways to hybrid quantum-classical supercomputing. But not all types of qubits allow for the quantum error correction needed to enable more reliable quantum computing. And without reliable quantum computing, valuable solutions to classically intractable problems are unlikely to be achieved. It’s essential to graduate from computing with noisy physical qubits to operating with reliable, logical ones.

One type of qubit that opens opportunities for more reliable quantum computation, with more and better logical qubits, is based on neutral atoms. Our partner, Atom Computing, uses neutral atoms as qubits to store and process quantum information through manipulation with pulses of light. Compared to other qubit technologies, neutral atoms have several advantages including the ability to be tightly packed in arrays while being held in place by lasers, as well as the ability to be moved around so that they can interact with other atoms to enable all-to-all connectivity. Neutral atoms also have low susceptibility to noise and the high fidelities needed for quantum error correction to work. Due to their lack of charge, neutral atoms can be kept only microns (millionths of a meter) apart. Thus, atomic arrays can accommodate large numbers of neutral atoms, each of which serves as a physical qubit, in a small amount of space, offering extensive scalability.

Atom Computing recently announced that they have achieved 99.6% two-qubit gate fidelity, which represents the highest fidelity of neutral-atom qubits in a commercial system and unlocks the ability to perform meaningful error correction. Qubits with such low error rates and all-to-all connectivity are prime candidates for Microsoft’s custom qubit-virtualization system, which creates logical qubits and enables the detection and correction of errors while performing reliable quantum computation.

Quantum Insider

Webinar with Dr. Krysta Svore and Dr. Ben Bloom on the offering from Microsoft and Atom Computing

Azure Quantum compute platform

To enable reliable quantum computing, Microsoft has developed the Azure Quantum compute platform. This compute platform combines Microsoft’s advanced qubit-virtualization system with quantum-processing units (QPUs) from our industry-leading hardware partners to create logical qubits by detecting and correcting the errors inherent in the QPU’s physical qubits, in turn enabling reliable quantum computation.

Microsoft’s qubit-virtualization system is a core component of the Azure Quantum compute platform and builds on Microsoft’s world-class quantum error correction innovations. It offers high performance quantum error correction, custom-tailored for the underlying QPU and can be applied to a variety of QPU types. By co-designing the quantum computer with Atom Computing, we are enabling faster, more advanced, and more efficient quantum error correction as part of our qubit-virtualization system.

Creating and entangling logical qubits 

To enable complex quantum computations, qubits must meet several criteria. The physical qubits must have low enough error rates to enable error detection and correction so that they can be used to create reliable logical qubits. Those logical qubits must then be able to not only be entangled, but also to undergo many logical operations without failure to enable successful solutions to complex computations. 

By applying Microsoft’s qubit-virtualization system to Atom Computing’s neutral-atom qubits, the teams created 24 logical qubits and entangled them in a cat state, or Greenberger-Horne-Zeilinger (GHZ) state—this represents the highest number of entangled logical qubits on record. Entanglement of the qubits is evidenced by their error rates being significantly below the 50% threshold for entanglement (Figure 1).

When using neutral atoms as qubits and attempting to hold them in place with lasers, some atoms can still become lost, resulting in a loss of that qubit and its quantum information. Thus, the teams took steps to not only detect errors but also to detect and correct losses of neutral-atom qubits during experiments. After creating logical qubits from neutral atoms and detecting errors and losses, the error rate in the logical qubits was 10.2%, which is 4.1 times better than the baseline physical error rate of 42% (Figure 1). When errors were detected and losses were detected and corrected, the error rate was 26.6%, which is 1.6 times better than the physical error rate (Figure 1). This is the first demonstration on record of loss correction in a commercial neutral-atom system.  

Twenty-eight logical qubits, created from 112 physical qubits, were used to perform successful computations based on the Bernstein-Vazirani algorithm. Furthermore, the logical qubits were able to produce a more accurate solution than the corresponding computation based on physical qubits. The ability to compute while detecting and correcting errors is a critical component to scaling to achieve scientific quantum advantage.

 Error rates of physical qubits (baseline) and the logical qubits created from them. The 24 logical qubits, which were entangled in a cat state, were created by applying Microsoft’s qubit-virtualization system to Atom Computing’s neutral atoms.

Figure 1. Error rates of physical qubits (baseline) and the logical qubits created from them. The 24 logical qubits, which were entangled in a cat state, were created by applying Microsoft’s qubit-virtualization system to Atom Computing’s neutral atoms. The entanglement of the logical qubits is evidenced by the error rate being below the 50% threshold (dashed line). The error rate of the logical qubits varied depending on whether losses were only detected or were detected and corrected. In both cases, the logical qubits demonstrated a higher fidelity of entanglement than the corresponding physical qubits.

Commercial offering from Microsoft and Atom Computing

Microsoft and Atom Computing are offering a state-of-the-art reliable quantum computer built from Atom’s neutral-atom hardware and Microsoft’s qubit-virtualization system, and that integrates with Azure Elements. This offering is a comprehensive scientific suite that combines logical qubits, cloud HPC, and advanced AI models that, together, enable exploration across multiple fields, including chemistry and materials science, while also providing opportunities for skilling and education.

Azure Elements, which is included in the offering, uses AI and cloud high-performance computing (HPC) to greatly accelerate the pace of scientific research and discovery. Capabilities of Azure Elements include Generative Chemistry, which brings the power of generative AI to chemistry research to enable the rapid discovery of molecules that are novel, synthesizable, and useful. In addition, Accelerated DFT offers substantial increases in speed over other DFT (density functional theory) codes for modeling quantum-mechanical properties. These and other user-friendly features of Azure Elements greatly increase the productivity of scientific research, especially when coupled with the quantum capabilities of the Azure Quantum compute platform. 

This suite—offered by Microsoft and Atom Computing—will enable not only exploration of quantum applications, but also the creation of new datasets by the quantum computer that may be used to train AI models for improved performance or accuracy. And while the offering targets scientific discovery, it’s also possible to explore how quantum capabilities may provide solutions to complex problems in other industries. 

“We are excited to continue our collaboration with Microsoft, which has already led to significant milestones in quantum computing. By coupling our state-of-the-art neutral-atom qubits with Microsoft’s qubit-virtualization system, we are now able to offer reliable logical qubits on a commercial quantum machine. This system will enable rapid progress in multiple fields including chemistry and materials science.” Dr. Ben Bloom, Founder and CEO of Atom Computing

Learn more about our latest achievements in quantum computing