Nvidia Launches AI Models to Tackle Quantum Errors
The open-source Ising models are designed to speed processor calibration and improve error-correction decoding in quantum systems.
Topics
[Source photo: Krishna Prasad/Fast Company Middle East]
NVIDIA has launched a family of open-source artificial intelligence models designed to help researchers and enterprises tackle two of quantum computing’s biggest engineering problems, processor calibration and error correction.
The model family, called Ising, is aimed at improving the reliability of quantum systems, whose qubits are highly sensitive to noise and other disruptions that can introduce errors during computation.
Quantum computers process information using qubits, which are highly sensitive and prone to disruption. Even small disturbances can lead to errors, making it difficult to scale these systems for real-world applications.
Nvidia said such challenges remain a major obstacle to scaling quantum machines for practical applications.
The chipmaker said one part of the system, Ising Calibration, uses a vision-language model to interpret measurements from quantum processors and automate continuous calibration, reducing the process from days to hours.
A second component, Ising Decoding, uses two variants of a 3D convolutional neural network, tuned for either speed or accuracy, to perform real-time decoding for quantum error correction.
According to the company, the decoding models are up to 2.5 times faster and three times more accurate than pyMatching, an open-source method widely used for quantum error-correction decoding.
“AI is essential to making quantum computing practical,” said Jensen Huang. “With Ising, AI becomes the control plane, the operating system of quantum machines, transforming fragile qubits to scalable and reliable quantum-GPU systems.”
Nvidia said the models are being adopted by a range of research institutions and quantum computing firms, including Academia Sinica, Fermi National Accelerator Laboratory, Harvard John A. Paulson School of Engineering and Applied Sciences, Infleqtion, IQM Quantum Computers, Lawrence Berkeley National Laboratory’s Advanced Quantum Testbed and the UK National Physical Laboratory.
The company said the tools are available on GitHub, Hugging Face and its Build developer hub, and can be run locally to help users retain control over data and infrastructure.
The launch adds to Nvidia’s push to position AI and accelerated computing as core infrastructure for quantum development.
The company cited an estimate from analyst firm Resonance that the quantum computing market could exceed $11 billion by 2030, though that growth will depend on progress in areas such as error correction and scalability.
