Insider Brief:

• NVIDIA positions quantum computing not as a hardware race but as a shared infrastructure challenge requiring accelerated computing, deep integration, and AI-driven collaboration.

• At GTC 2025, NVIDIA reinforced this role by announcing the NVIDIA Accelerated Quantum Center and announcing a series of quantum partnerships that emphasize hybrid quantum-classical systems.

• Companies such as QC Design, Pasqal, and SEEQC integrated NVIDIA’s CUDA-Q platform to address quantum error correction, fault tolerance, and simulation scalability.

• Research institutions and startups are using NVIDIA’s AI supercomputing to power quantum developments in imaging, layout optimization, error decoding, and hybrid algorithm development.

• Image Credit: NVIDIA

Quantum computing isn’t a new frontier for NVIDIA—it’s a systems-level challenge, and one the company is uniquely positioned to support. NVIDIA doesn’t see quantum as a race to build the biggest machine, but as a shared infrastructure problem, and one that demands accelerated computing, scaled collaboration, and deep integration across disciplines.

This mindset was on quiet display at this year’s GTC. While the headlines were dominated by open-source robotics, next-gen GPUs, and countless AI infrastructure reveals, tucked between the flashier days of AI and automation was Quantum Day, which humbly delivered a telling story about NVIDIA’s long-term direction.

Over the course of the week, a steady stream of announcements—some from the main stage, others in parallel—made clear that NVIDIA is not entering the quantum hardware race. It’s doing what it has always done: building the tools and systems that will help others go further, faster.

This approach is nothing new. To understand where NVIDIA fits into quantum’s future, you only have to look at how they’ve approached robotics and autonomous vehicles. As Sam Stanwyck, Group Product Manager for Quantum Computing at NVIDIA, put it in a recent interview with the Quantum Insider:

“We don’t build our own self-driving car, but we help everyone else who does. We don’t build our own robots, but we help everyone else who does. We don’t build our own quantum computer, but our mission is to bring AI and accelerated computing to help everyone else who does.”

That’s the company’s MO. NVIDIA excels in AI and accelerated computing—then embeds that power into the broader ecosystem. As Stanwyck emphasized, “We are an accelerated computing company, and we see quantum as an important part of the future of accelerated computing.” The goal isn’t to dominate the quantum hardware race but to accelerate it—by reducing bottlenecks, speeding up error correction, and enabling hybrid quantum-classical workflows that actually scale.

This strategy is embodied in the development of the NVIDIA Accelerated Quantum Center, where NVIDIA’s supercomputing hardware will sit alongside QPUs from industry partners and research institutions. The aim is to explore how the “holy trinity”—compute, AI, and quantum—can operate not in silos, but in synergy. As Stanwyck made clear, “Quantum is very important to us… and we also see that accelerated computing and AI are going to be essential to get quantum computing to where we all want it to be.”

In addition to NVAQC, a wave of quantum-related partnerships and integrations emerged throughout the week—some announced during GTC, others announced in parallel—all pointing to NVIDIA’s growing role as a central enabler of the quantum ecosystem. Here’s a breakdown of every quantum-powered announcement from GTC 2025 and the surrounding days:

NVIDIA Accelerated Quantum Research Center

NVIDIA is launching the Accelerated Quantum Research Center in Boston to integrate quantum hardware with AI supercomputers in an effort to advance practical quantum computing. The center will collaborate with partners including Quantinuum, QuEra, Quantum Machines and top universities such as Harvard and MIT to address challenges like qubit noise and error correction. It will use NVIDIA’s GB200 NVL72 systems and CUDA-Q platform to develop hybrid quantum algorithms and AI-driven quantum applications. NVAQC is set to begin operations later this year. Read more here.

QC Design Pioneers GPU-Accelerated Quantum Fault-Tolerance Design

QC Design has integrated NVIDIA’s cuQuantum SDK into its Plaquette software, enabling GPU-accelerated, full-state simulations of fault-tolerant quantum circuits. This allows researchers to simulate over 400 qubits on a single RTX 4000 GPU with 20GB memory—far beyond the 60-qubit limit of CPU-based simulators. Plaquette now achieves up to 180x faster sampling speeds for 60-qubit circuits and models over 20 types of hardware imperfections. Read more here.

Quantum Machines Announces NVIDIA DGX Quantum Early Access Program

Quantum Machines has launched the NVIDIA DGX Quantum Early Customer Program, introducing a tightly integrated quantum-classical system that combines its OPX1000 quantum control platform with NVIDIA’s GH200 Grace Hopper Superchips. The system achieves real-time quantum error correction and AI-driven calibration with latencies under 4 microseconds. Early adopters include MIT’s EQuS group, the Israeli Quantum Computing Center, Diraq, and ENS Lyon, using the platform for applications like hybrid algorithms and fast feedback. Read more here.

Pasqal to Advance Hybrid Quantum Computing with NVIDIA CUDA-Q Platform

Pasqal has integrated its neutral-atom quantum computing platform with NVIDIA’s CUDA-Q, enabling seamless hybrid quantum-classical programming across CPUs, GPUs, and QPUs. This collaboration expands Pasqal’s developer tools by combining its Pulser library with CUDA-Q’s Python and C++ interfaces, supporting advanced analog quantum programming and simulations. The integration opens new workflows for the HPC community, enhancing interoperability and accelerating quantum application development. Read more here.

SEEQC Develops Digital Interface for Real-Time Quantum-Classical Integration with NVIDIA-Powered Error Correction

SEEQC and NVIDIA have demonstrated the first fully digital, chip-to-chip interface between a quantum processor and a GPU, enabling quantum error correction with microsecond latency and 1000x less bandwidth. Powered by SEEQC’s Single Flux Quantum (SFQ) technology and integrated with NVIDIA’s CUDA-Q platform, the interface removes key scaling bottlenecks in quantum computing. This is a notable development toward heterogeneous computing by allowing real-time, low-latency communication between quantum and classical systems. Read more here.

MITRE Builds New Quantum Imaging Using NVIDIA CUDA-Q

MITRE and NVIDIA are partnering to accelerate simulations for designing quantum imaging systems, including MITRE’s Walsh Imaging technology. Walsh Imaging can noninvasively capture nanoscale electromagnetic signals from semiconductors or brain neurons in real time, offering breakthroughs in medicine, microelectronics, and security. By leveraging NVIDIA’s CUDA-Q platform and DGX SuperPOD, MITRE can simulate and optimize these complex quantum systems in under an hour, a task that previously took days. The collaboration highlights the growing role of GPU-accelerated computing in advancing quantum sensing and imaging technologies. Read more here.

Quantum Rings Now Available for NVIDIA CUDA-Q, Streamlining Quantum Simulation

Quantum Rings has integrated its high-performance quantum circuit simulation technology with NVIDIA’s CUDA-Q platform, enabling GPU-accelerated simulations of large-scale quantum circuits. This allows researchers and developers to rapidly iterate on complex quantum algorithms using both consumer GPUs and HPC clusters. The integration supports faster, more cost-effective testing in preparation for future fault-tolerant quantum hardware. Now available as a standard simulator in CUDA-Q, Quantum Rings expands access to quantum simulation tools for both academia and industry. Read more here.

Q-CTRL Accelerating quantum advantage by scaling error suppression with NVIDIA and OQC

Q-CTRL, in partnership with NVIDIA and Oxford Quantum Circuits, has achieved a reduction in compute costs for quantum error suppression by accelerating layout ranking with NVIDIA GPUs. Their software, Fire Opal, uses AI-driven techniques to map quantum circuits to hardware efficiently—a process that becomes increasingly complex as qubit counts grow. GPU acceleration using NVIDIA’s RAPIDS and cuDF libraries reduced layout selection times, achieving speedups over CPU-based methods in large-scale benchmarks. These advances not only cut costs and execution times but also improve algorithm fidelity and scalability. Read more here.

NVIDIA and QuEra Decode Quantum Errors with AI

NVIDIA and QuEra have developed a transformer-based AI decoder for quantum error correction, outperforming traditional decoders like maximum likelihood estimation while offering greater scalability. Trained using GPU-accelerated simulations via CUDA-Q and validated with data from QuEra’s neutral-atom QPU, the decoder enables faster and more efficient decoding of magic state distillation circuits—critical for fault-tolerant quantum computing. The AI model achieves higher fidelity at greater acceptance ratios and completes decoding in under a millisecond, compared to over 100 ms for MLE. Scaling the decoder to higher code distances will leverage AI supercomputers like NVIDIA’s NVAQC and Eos to generate massive training datasets and support real-time decoding for practical quantum systems. Read more here.

Infleqtion Announces Contextual Machine Learning to Power AI Developments with NVIDIA CUDA-Q and Quantum-Inspired Algorithms

At GTC 2025, Infleqtion unveiled Contextual Machine Learning, an AI approach designed to process data from multiple sources over extended timeframes for improved real-time decision-making. Implemented on NVIDIA A100 GPUs using the CUDA-Q platform, CML enhances AI performance in defense, energy, and autonomous systems while laying the foundation for future quantum-powered machine learning. This work builds on Infleqtion’s prior CUDA-Q-driven breakthroughs in quantum materials design and highlights the growing convergence of AI supercomputing and quantum computing. Read more here.