NVIDIA has introduced Ising, the world’s first family of open-source AI models designed specifically to accelerate the development of practical quantum computing. The Ising models deliver industry-leading performance in quantum processor calibration and error correction, achieving results up to 2.5 times faster and three times more accurate than traditional methods.

The new model family is already being adopted by leading quantum companies, research institutions, and academic organizations, including Academia Sinica, Fermi National Accelerator Laboratory, Harvard’s John A. Paulson School of Engineering and Applied Sciences, IQM Quantum Computers, and the U.K. National Physical Laboratory.

Advancing Practical Quantum Computing

Building useful, large-scale quantum systems requires major advances in two key areas: processor calibration and error correction. NVIDIA positions AI as the critical tool to overcome these challenges, enabling today’s fragile quantum systems to evolve into reliable, scalable computing platforms.

The Ising models are named after a foundational mathematical framework that helped simplify complex physical systems. In a similar spirit, NVIDIA’s Ising platform provides powerful, scalable AI tools to address the most difficult problems in hybrid quantum-classical computing.

Breakthrough Performance

Ising models significantly enhance the efficiency of quantum systems by improving both calibration and error correction processes. They enable researchers to handle more complex computations while maintaining higher accuracy and speed.

“AI is essential to making quantum computing practical,” said Jensen Huang, founder and CEO of NVIDIA. “With Ising, AI effectively becomes the control layer — the operating system — for quantum machines, helping transform delicate qubits into scalable and reliable systems.”

Key Capabilities

The NVIDIA Ising platform includes advanced models and tools designed to accelerate quantum development:

• Ising Calibration: A vision-language AI model that interprets quantum processor measurements in real time, enabling automated calibration processes that reduce setup times from days to hours.

• Ising Decoding: Advanced neural network models optimized for either speed or accuracy, delivering real-time error correction that significantly outperforms current open-source standards.

Growing Ecosystem Adoption

A broad range of organizations is already implementing Ising in their quantum research and development efforts. Institutions such as Cornell University, Sandia National Laboratories, UC San Diego, and the University of Chicago are using the platform to enhance quantum error correction and system performance.

Tools for Developers

NVIDIA is also providing a comprehensive toolkit, including workflow guides, training datasets, and NVIDIA NIM™ microservices. These resources allow developers to customize models for specific hardware and use cases with minimal setup. Importantly, the models can run locally, ensuring data privacy and security.

Powering the Future of Quantum Systems

Ising integrates seamlessly with NVIDIA’s broader quantum ecosystem, including the CUDA-Q™ platform for hybrid computing and NVQLink™ technology, which enables real-time communication between quantum processors and GPUs.

With the global quantum computing market projected to exceed $11 billion by 2030, NVIDIA’s Ising platform represents a major step toward making quantum computing practical, scalable, and widely accessible.