NVIDIA has unveiled a new AI model that will accelerate the commercialization of quantum computing.

On the 16th, NVIDIA announced 'NVIDIA Ising,' the world's first open-source quantum AI model family, paving the way for innovative improvements in quantum processor performance and error correction technology.

Ising is named after a mathematical model that simplifies and interprets complex physical systems.

This product family focuses on accelerating quantum calibration and error correction using AI, which are core technologies required for building hybrid quantum-classical systems.

Through this, NVIDIA supports researchers and companies in implementing practical quantum applications.

NVIDIA Izing consists of two main components.

First, Ising Calibration utilizes a Vision Language Model (VLM) to rapidly analyze measurement data from quantum processors and automatically correct it.

It significantly increases research efficiency by allowing tasks that previously took several days to be completed within a few hours.

Ising Decoding performs real-time decoding for quantum error correction based on a 3D Convolutional Neural Network (CNN).r />
It is available in two versions, speed-oriented and accuracy-oriented, and achieves up to 2.5 times faster and 3 times higher accuracy than the existing open source tool pyMatching.

NVIDIA founder and CEO Jensen Huang stated, “AI is a key technology for making quantum computing practical,” adding, “I-Jing acts as the control layer of quantum machines, transforming vulnerable qubits into a trusted quantum-GPU system.”

Currently, Aijing is being adopted by major research institutions and companies around the world, including Academia Sinica in Taiwan, Harvard Engineering, Fermi National Accelerator Laboratory, IQM Quantum Computer, Lawrence Berkeley National Laboratory, and the National Physical Laboratory in the UK.

In addition, research utilizing the Ising decoding model is underway in academia, including at Cornell University, UC Santa Barbara, the University of Chicago, and Yonsei University.

NVIDIA added Izing to its open model portfolio.

This portfolio includes Nemotron for AI agents, Cosmos for physical simulation, Alphamayo for autonomous driving, Isaac GR00T for robotics, and BioNeMo for life science research.

All models are accessible on GitHub, Hugging Face, and build.nvidia.com.

Market research firm Resonance projected that the global quantum computing market size will exceed $11 billion by 2030.

Amidst this growth, NVIDIA's Izing is being evaluated as a practical tool capable of solving quantum error correction and scalability issues.