September 10, 2026 -- Quantum computing and artificial intelligence are rapidly converging, and NVIDIA's latest Ising Calibration 1.5 release marks another significant step towards autonomous quantum computing.

Designed specifically for quantum processor calibration, NVIDIA Ising is an open-source family of AI models that spans key quantum workloads. Ising 1.5 combines advanced vision-language capabilities with enhanced in-context learning, enabling it to interpret unfamiliar diagnostic data, recommend hardware tuning and automate complex calibration tasks without requiring extensive retraining. The model is now available in a highly efficient quantised version that runs on a single GPU or NVIDIA DGX Spark, making AI-driven quantum control more accessible than ever.

As one of the first companies to integrate NVIDIA's open-source Ising models into the operation of a live photonic quantum computer, Aegiq has deployed AI-enhanced calibration on Artemis, its first-generation system at the UK's National Quantum Computing Centre (NQCC). Running locally on an NVIDIA DGX Spark, the model regularly optimises the performance of Aegiq's quantum dot photon sources by balancing critical parameters including brightness, purity and indistinguishability.

The results have been immediate. AI-assisted calibration has reduced weekly engineering effort by 3×, enabling non-specialists to perform routine optimisation while freeing quantum engineers to focus on developing the next generation of photonic quantum technologies.

This not only provides a productivity improvement, but it is also a prerequisite for scaling quantum systems.  Today's systems contain hundreds of components; tomorrow's fault-tolerant quantum computers will contain millions. Manual calibration will not be able to scale. AI-powered, agent-based workflows, such as NVIDIA Ising, provide the foundation for autonomous operation, increased uptime and significantly lower operational costs.

The release of Ising 1.5 supports this direction. With stronger zero-shot reasoning, with a potential of over 86% improvement in in-context learning, for calibration tasks, over its predecessor and efficient deployment on compact GPU platforms, helping the transition towards usable quantum infrastructure.

For Aegiq, AI extends beyond calibration. The company is also exploring NVIDIA's quantum platform to accelerate quantum error correction workflows alongside its QGATE architecture, creating a high performing combination of AI, GPU acceleration and photonic quantum computing.

Aegiq is using NVIDIA Ising demonstrating how AI is becoming an essential operating layer for practical quantum computers by automating complex engineering tasks to enable scalable, fault-tolerant quantum systems.