NVIDIA Ising Enables Fully Automated Quantum Computer Calibration with Enhanced In-Context Learning | NVIDIA Technical Blog

Introduces NVIDIA Ising Calibration 1.5, a 31B-parameter vision language model for automated quantum computer calibration. Key contributions include enhanced in-context learning for interpreting unfamiliar diagnostic results, an NVFP4-quantized version for single-GPU deployment, training on diverse qubit modalities, and evaluation on the QCalEval benchmark demonstrating state-of-the-art zero-shot and in-context learning performance. The post also provides open-source model weights, datasets, and deployment blueprints.