Nord Quantique: Multimode Quantum Encoding Cuts Qubit Count & Boosts Error Correction Performance

Nord Quantique Quantique, a quantum computing firm based in Sherbrooke, Canada, has demonstrated a new approach to quantum error correction (QEC) utilising multimode encoding with bosonic qubits and the Tesseract code. This technique aims to reduce the number of physical qubits required for QEC, potentially delivering smaller, more energy-efficient and scalable quantum systems. The company reports successful detection of multiple error types, including bit flips, phase flips, and leakage errors, achieving stable performance over 32 error correction cycles. Nord Quantique anticipates delivering utility-scale quantum computers with over one hundred logical qubits by 2029, and projects significant reductions in energy consumption – for example, solving RSA-830 in one hour using 120 kWh compared to classical HPC estimates of 280,000 kWh over nine days.

Nord Quantique has demonstrated a novel approach to quantum error correction utilising multimode encoding with the Tesseract code. This technique encodes individual qubits across multiple quantum modes – distinct resonance frequencies within an aluminium cavity – providing inherent redundancy and enhanced protection against errors. Crucially, this implementation extends beyond simple bit-flip and phase-flip correction to also detect leakage errors, where the qubit escapes the encoding space, a common limitation of single-mode systems. Nord Quantique anticipates delivering utility-scale quantum computers with over one hundred logical qubits by 2029, paving the way for practical applications of quantum computation.

Researchers at Nord Quantique demonstrated the efficacy of the Tesseract code through experiments showing stable quantum information across 32 error correction cycles, discarding only 12.6% of data during post-selection filtering of imperfect runs. Increasing the number of modes further amplifies error detection capabilities, suggesting a pathway to improved quantum error correction performance as the system scales. This contrasts with conventional approaches that require a substantial increase in physical qubits to achieve comparable error mitigation, offering a potentially more efficient architecture.

A key advantage of multimode encoding lies in its potential to reduce the physical qubit overhead associated with quantum error correction. The company anticipates a near 1:1 ratio of physical cavities to logical qubits as systems expand, representing a significant improvement over many current designs. This approach also mitigates auxiliary decay errors, suppresses silent errors, and facilitates the extraction of confidence information to refine error detection and correction strategies, enhancing overall system reliability.

From a practical standpoint, Nord Quantique projects a quantum computer with 1,000+ logical qubits will occupy approximately 20 square metres, making it suitable for integration within existing data centre infrastructure. Performance estimates, using RSA-830 as a benchmark, suggest solving the problem in one hour at 1 MHz, consuming 120 kWh, a substantial reduction compared to the 1,300 kW over nine days and 280,000 kWh required by classical HPC. These figures also present a favourable comparison to alternative quantum computing methodologies, highlighting the potential for energy-efficient computation.