Nvidia Launches First AI Models for Quantum Computing

Nvidia has unveiled Ising, the world's first open AI model family designed for quantum computing calibration and error correction. The two models address the core challenge of scaling quantum computers, where fragile, error-prone qubits require constant real-time correction and tuning. Ising Decoding uses 3D convolutional neural networks to outperform the current industry standard by 2.5 times in speed and three times in accuracy. Ising Calibration uses a vision-language model to automate hardware tuning. Both are already being adopted by institutions including Cornell University, Sandia National Laboratories, IonQ and Q-CTRL.