Infleqtion cuts qubit needs with new error correction code

Infleqtion researchers have achieved an approximately six-to-one ratio of physical to logical qubits using a new quantum error correction code, a fivefold improvement over comparable surface-code methods, the company says. The company integrated its qLDPC library, open-sourced in 2023, with NVIDIA CUDA-Q Logical to accelerate development of this high-rate code, packing more logical information into existing quantum hardware.

“Quantum software can change how much hardware we need to reach utility scale,” says Pranav Gokhale, Chief Technology Officer and General Manager of Quantum Computing at Infleqtion, emphasizing the importance of co-designing code and machine for optimal performance.

qLDPC Integration Reduces Physical Qubit Needs for Error Correction

A ratio of approximately six physical data qubits per logical qubit was recently achieved by Infleqtion researchers, a substantial improvement over existing surface-code approaches that typically require significantly more physical qubits to maintain data integrity. This advancement stems from the integration of Infleqtion’s open-source qLDPC library with the NVIDIA CUDA-Q Logical software platform, accelerating the development of high-rate quantum error correction codes, according to the company.

The resulting code demonstrates a marked reduction in the physical resources needed to represent and protect quantum information, bringing practical quantum computation closer to reality. This lower ratio is critical because the sheer number of physical qubits required for effective error correction has long been a primary obstacle to building useful quantum computers. The successful construction and validation of this code relied on a collaborative effort, using the strengths of both Infleqtion’s qLDPC library and NVIDIA’s CUDA-Q Logical.

Infleqtion initially open-sourced qLDPC in 2023, working alongside researchers from JPMorgan Chase, and has since expanded the workflow to include syndrome extraction, noise modeling specific to neutral atoms, and decoder benchmarking. This builds upon the company’s previous work utilizing the NVIDIA Ising family of models for real-time decoding, demonstrating a commitment to a full-stack approach, the company says. Portions of this integration were also developed with the aid of artificial intelligence, a practice Infleqtion credits with enhancing the efficiency of its logical-qubit operations.

The company’s 100-qubit Sqale system, running at the UK’s National Quantum Computing Centre, provides a testbed for these advancements.

The achievement of a physical-to-logical qubit ratio below 10:1 signifies progress toward Infleqtion’s architectural goals for near-term quantum systems. Better error correction means more computational power from fewer physical qubits.” This integrated approach is not merely about reducing qubit counts; it’s about maximizing the computational potential of existing hardware. The ability to pack more logical information into a given quantum processor is a key step toward overcoming the limitations of current technology.

Infleqtion presented details of this work at IEEE Quantum Week 2026 in Toronto, including a tutorial on accelerated decoders for quantum error correction presented by Pranav Gokhale and a paper session featuring “Hardware-Aware Optimization of Echoed-Conditional Displacement and Rotation Sequence Parameters.” The company’s broader roadmap includes targets of more than 30 logical qubits in 2026 and more than 100 by 2028, ultimately aiming for a fault-tolerant system with 1,000 logical qubits by 2030.

Sam Stanwyck, Director of Quantum Product at NVIDIA, emphasized the importance of this development, stating, “The path to useful quantum computing depends on turning advances in error correction into full-stack architectures that can run on real hardware.”

CUDA-Q Logical Validates High-Rate Logical Qubit Construction

The team validated this high-rate code using an early-access version of NVIDIA’s CUDA-Q Logical software platform, confirming its ability to support multiple distinct logical qubits within a single encoded structure. This collaborative effort addresses a critical bottleneck in scaling quantum computers, the substantial physical qubit overhead traditionally demanded by error correction schemes. Infleqtion’s approach, built on reconfigurable neutral-atom arrays, uses a platform well-suited to high-rate codes due to its parallel operations and flexible connectivity.

The company is using artificial intelligence to further enhance the efficiency of its logical-qubit operations, a practice that has already yielded promising results. The company, listed on the NYSE as INFQ with a headcount of about 250 people and total funding exceeding $550 million, is positioned as a key player in the rapidly evolving quantum landscape.

Infleqtion Advances Workflow with Noise Modeling & AI Assistance

The workflow extends beyond code development, now incorporating real-time noise modeling and decoder benchmarking, building on Infleqtion’s prior work with NVIDIA’s Ising models for real-time decoding. This expansion toward physical execution allows for a more complete evaluation of logical qubit performance under realistic conditions. The tutorial, led by Pranav Gokhale, explored GPU-enabled algorithmic and AI decoding techniques, reflecting the increasing importance of hardware acceleration in quantum computing. The company’s approach uses reconfigurable neutral-atom arrays, a platform well-suited to high-fidelity qubit control and scalability.

Infleqtion’s systems are running at the UK’s National Quantum Computing Centre, demonstrating the growing demand for quantum computing solutions across diverse sectors, the company states. Beyond government applications, Infleqtion has also collaborated with NVIDIA to publish the first demonstration of a materials science application utilizing logical qubits, showcasing the potential of fault-tolerant quantum computing to address real-world problems.

More like this

Quantum Computing Business NewsQuEra Computing finds nearly half of firms want quantum fault-tolerance plans

Quantum ApplicationsZapata Quantum launches software to map business needs to quantum hardware

Quantum Computing Business NewsQC Design’s Meridian AI designs quantum circuits better than experts

Quantum Research NewsDepartment of Energy SCAC report charts path to error-corrected quantum computers