Diraq and QM Integrate AI with Quantum Computing Using NVIDIA DGX Quantum

Diraq and QM Technologies have announced a groundbreaking integration of GPUs and silicon quantum processors, utilizing the NVIDIA DGX Quantum architecture. This collaboration has resulted in real-time communication speeds of 3.3 microseconds between quantum hardware and NVIDIA Grace-Hopper superchips, as announced in a press release.

Engineers at Diraq's Sydney laboratories demonstrated how GPU acceleration can address critical bottlenecks in quantum computing. They implemented three key applications: real-time readout enhancement, automated calibration using machine learning, and accelerated quantum state initialization. These advancements were achieved within a week of installing the DGX Quantum system, showcasing its immediate impact on quantum computing research.

The integration of NVIDIA Grace Hopper with QM's OPX1000 control system allows quantum processors to leverage AI and GPU-accelerated computation in real-time. This setup enables complex signal processing tasks and automated calibration processes that traditionally required extensive manual tuning.

The collaboration exemplifies the potential of combining classical computing power with quantum computing, paving the way for future advancements in hybrid quantum-classical architectures. The results will be highlighted at GTC Paris 2025, where further discussions on hybrid algorithm development will take place.