Nvidia releases open AI models for quantum computing tasks — 'Ising' said to be 2.5x faster and 3x more accurate than existing tools for decoding

Nvidia has unveiled Ising, a set of open-source AI models tailored for quantum processor calibration and real-time error correction decoding. These models aim to address key challenges in quantum computing, such as calibration and decoding bottlenecks. Ising Calibration, a vision-language model, reduces calibration time significantly, while the Ising Decoding family offers faster and more accurate decoding for quantum error correction. Nvidia's Ising models have shown significant improvements over existing tools, attracting adopters like Fermilab, Harvard, and IonQ. Despite being open-source, Ising relies on Nvidia's proprietary software stack for optimal performance.