Artificial Intelligence, Quantum Error Correction

Riverlane and Nvidia speed up quantum error correction with AI

Avatar of Ivy Delaney
October 14, 2026 by Ivy Delaney

Riverlane is collaborating with Nvidia to address a critical challenge in quantum computing: maintaining the accuracy of error correction as hardware inevitably drifts. Future quantum error correction systems may require processing terabytes of data per second, and the partnership focuses on accelerating the calibration of decoder “weights”, numerical settings that reflect the behavior of unstable quantum hardware.

The companies are exploring how Nvidia’s H100 GPU can speed up this calibration process, allowing quantum systems to adapt to changing qubit frequencies and defects. This approach, Riverlane reports, has already reduced a training cycle for surface code from three minutes on a CPU to just 2.8 seconds.

NVIDIA H100 GPU Accelerates Quantum Error Correction

Changes in quantum hardware necessitate frequent recalibration of error correction decoders, a process Riverlane is now accelerating with assistance from Nvidia’s accelerated computing platforms. The decoder’s “weights”, derived from measurements of quantum hardware, must be constantly adjusted because of inherent instabilities and evolving defects within the system. This dynamic behavior demands a calibration process fast enough to keep pace with the quantum computer itself, a challenge the collaboration directly addresses.

“Riverlane is using AI to help error correction keep pace with changing quantum hardware,” said Sam Stanwyck, Director of Quantum Product at NVIDIA. Accelerating this learning process allows for more frequent calibration cycles, potentially freeing researchers from extensive tuning and enabling greater focus on scientific discovery. Recalibrating the decoder is framed as a learning problem, and Nvidia’s hardware is intended to make that learning routine.

The partnership investigates how advances in quantum computing, artificial intelligence, and accelerated computing can converge to support the development of quantum systems. Riverlane will evaluate AI-enhanced approaches to error correction, exploring how different computing technologies can support future quantum processors and how its QEC technologies could integrate with Nvidia’s accelerated computing and quantum platforms. This broader investigation extends beyond immediate speed gains to consider the future of quantum-classical computing architectures.

Steve Brierley, founder and CEO of Riverlane, emphasizes the foundational role of quantum error correction. “Quantum error correction is the key technology that will enable useful quantum computers,” Brierley states. “As quantum systems scale, QEC becomes one of the world’s most demanding real-time computing challenges.” The collaboration aims to address this scaling challenge by using Nvidia’s capabilities to process the terabytes of data per second that future QEC systems may require for error-free computation.

Riverlane uses NVIDIA H100 GPU to test how well different decoder settings perform. It’s about transforming how quantum computers are built and operated. Riverlane will continue to investigate how Nvidia accelerated computing can make quantum error correction faster, more adaptive, and more practical for utility-scale quantum computers, ultimately opening the way to these machines to reach their full potential. The companies are focused on enabling more frequent calibration, which allows researchers to spend less time tuning their quantum computers and more time conducting research.

Quantum error correction is the key technology that will enable useful quantum computers.

Steve Brierley, founder and CEO of Riverlane

Faster Calibration Enables Routine Quantum System Updates

The calibration of quantum error correction decoders, once a multi-minute process, now occurs in under three seconds thanks to a new approach by Riverlane using Nvidia’s H100 GPU. This dramatic reduction in calibration time stems from an AI-based system that learns optimal decoder settings directly from experimental data, eliminating the need for extensive manual tuning. Riverlane’s Local Clustering Decoder (LCD), adapted for high-throughput processing on the GPU, achieves this speed by rapidly testing and refining decoder settings against real-time data from quantum hardware.

The instability of quantum hardware necessitates frequent recalibration; frequencies shift and defects change over time, demanding constant adjustments to error correction protocols. The H100 GPU’s processing power allows for more than four million surface code syndromes to be decoded per second, a substantial increase from the approximately 59,000 syndromes processed per second on a single CPU core.

This accelerated processing isn’t simply about speed, but about enabling a continuous learning loop for the decoder, improving its accuracy and adaptability, Nvidia says. The system begins with an initial estimate of error patterns, then uses the GPU to quickly evaluate and refine decoder settings based on data from the quantum hardware itself. This approach fundamentally alters the calibration process, transforming it from a time-consuming manual task into a routine, automated procedure.

The ultimate goal is to enable more frequent and efficient calibration. Rapid adaptation to hardware changes is important as quantum systems scale, and this collaboration represents a step toward building truly useful and scalable quantum computers.

Accelerated computing is becoming the engine of useful quantum computing, transforming how quantum computers are built and operated.

Sam Stanwyck, Director of Quantum Product at NVIDIA
More like this
Quantum HardwareAlice & Bob tests quantum decoding with Quantum Machines
Quantum Error CorrectionHyperbolic codes could quadruple efficiency of quantum error correction
Quantum Error CorrectionGrover’s search runs on three logical qubits with this new code
Technology NewsInside D-Wave’s Dual-Rail Qubit Roadmap and the New Haven Bet on Logical Qubits
Stay current

See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.

Tags:
error correction NVIDIA Quantum Computing quantum hardware Riverlane
Avatar of Ivy Delaney

Ivy Delaney

Ivy Delaney has been working with neural networks and machine learning since the mid-nineties, back when a couple of hidden layers and a long afternoon of training counted as ambitious. She has watched the field go from academic curiosity to the thing quietly running underneath everything, and she brings that long view to quantum computing. For Quantum Zeitgeist she covers the ground where the two fields meet. That means quantum machine learning and the variational algorithms it leans on, and it also means the less glamorous but more interesting story of classical machine learning already doing real work inside quantum machines, decoding error-correcting codes, calibrating noisy hardware and learning the error models that simulators depend on. She writes about the hardware those algorithms have to run on too, and about the post-quantum cryptography scramble that the same hardware has set off. Her stories typically start with the paper, whether that is peer-reviewed work, conference proceedings or an arXiv preprint, with the source linked so you can hold a claim up against the research it came from. She is unimpressed by benchmarks that will not say what they beat, and by demonstrations that only work in the press release.

Latest Posts by Ivy Delaney: