Diraq Integrates NVIDIA Grace Hopper and NVQLink to Advance Utility-Scale Silicon Quantum Computing

Real-Time Latency Achievement: Diraq achieves approximately 3.3 µs round-trip latency between its silicon spin qubits and the NVIDIA GH200 Superchip using NVQLink for closed-loop quantum-classical control.Automated Calibration at Scale: NVIDIA Ising models automate charge stability map analysis, reducing expert physicist calibration time from one year of manual effort to model training completed in days.Hybrid Workflow Enablement: The integration supports real-time adaptive experiments and aligns Diraq’s compact, semiconductor-fabricated quantum processors with existing AI and HPC data center infrastructure for hybrid deployments.Diraq has integrated its silicon spin qubit processors with the NVIDIA GH200 Grace Hopper Superchip through the NVIDIA NVQLink interconnect, achieving round-trip latencies of approximately 3.3 microseconds. This architecture enables the real-time classical feedback, automated calibration, and adaptive control loops required to scale silicon quantum systems toward utility-scale operation.Diraq fabricates its quantum processors by modifying standard silicon transistors to encode quantum information, allowing millions of qubits on a single chip using mature semiconductor manufacturing processes. These silicon spin qubits operate on nanosecond timescales, creating a requirement for classical compute resources that can match this speed to orchestrate control sequences, analyze measurement data in real time, and close adaptive feedback loops for error correction and calibration.To meet these requirements, Diraq deployed a hybrid platform connecting its quantum processors to the NVIDIA GH200 Grace Hopper Superchip via NVQLink. The GH200 Superchip was integrated into Diraq’s system in May, 2025, with three real-time applications addressing core scaling challenges demonstrated within one week. NVIDIA CUDA-Q serves as the unified programming model for orchestrating quantum experiments with classical processing across the NVQLink interconnect.Diraq further incorporates the NVIDIA Ising family of open models to automate qubit calibration. GPU-accelerated, vision-capable machine learning models analyze charge stability maps and identify functional qubits through automated image analysis, eliminating the need for manual expert labeling of multi-dimensional tuning data.Key technical parameters of the implementation include:• Round-trip communication latency of approximately 3.3 µs between the quantum control system and classical compute resources.• Tight integration of GPU-accelerated machine learning models with live quantum experiments for real-time calibration signal generation.• Support for closed-loop adaptive experiments in which measurement results directly inform immediate adjustments to experimental parameters during runtime.The platform enhances Diraq’s internal development workflows while advancing the company’s strategy of deploying compact, energy-efficient quantum processors directly into standard data center environments alongside AI and HPC systems.Prior to the integration, manual calibration of silicon spin qubits required a full year of an expert physicist’s time, and offline measurement analysis delayed experiment redesign by multiple days. With GPU-accelerated models running on the GH200 and coupled through NVQLink, calibration model training now completes in days. Real-time feedback enables the system to adapt experiments dynamically as they execute, shifting team effort from repetitive tuning and offline analysis toward scalable architectures and error-correction pathway development.“NVIDIA is fundamentally changing our route toward utility-scale quantum computing, enabling real-time feedback and high-throughput automation across our stack. NVIDIA and Diraq are ideally suited to each other too — we’re the quantum and classical sides of the same coin in terms of being cost efficient and widely deployable,” said Andrew Dzurak, CEO and Founder of Diraq. NVIDIA is fundamentally changing our route toward utility-scale quantum computing, enabling real-time feedback and high-throughput automation across our stack. NVIDIA and Diraq are ideally suited to each other too — we're the quantum and classical sides of the same coin in terms of being cost efficient and widely deployable. Alignment with NVIDIA’s ecosystem provides Diraq access to mature software stacks, extensive documentation, and a large existing developer community already operating in data centers. Such compatibility reduces integration friction for hybrid quantum-classical workloads and supports Diraq’s core approach of leveraging established semiconductor infrastructure for high-density qubit scaling. Diraq plans to expand the platform across additional workflows, with future potential for its QPUs to generate training data that further improves classical AI models.The collaboration reflects the industry’s progression toward tightly coupled hybrid systems capable of delivering practical value at manageable infrastructure cost. Diraq’s silicon-based architecture enables dense on-chip integration and straightforward co-location with GPUs as accelerators, avoiding the bespoke facilities and high energy overhead that have historically constrained broader quantum adoption.Find out more here.Further articles, reports, and the latest quantum computing news may be found at The Qubit Report.EPB Quantum will host the inaugural Quantum in Business Conference on October 1, 2026, at The Westin Chattanooga. The event is designed to help companies The quantum sector accelerated from research to real deployments and policy mandates in mid-June 2026. The UK and Japan sealed a $24 billion partnership while Central New Mexico Community College hosted a Quantum Educators Workshop in early June 2026, introducing its Quantum Technician Bootcamp to 45 faculty members from across Sign up to receive our newsletter and other reports.We keep your data private and share your data only with third parties that make this service possible. Read our privacy policy for more info.Check your inbox or spam folder to confirm your subscription.