Altera’s design links QPU to NVIDIA GPU with NVQLink

Altera Corporation, in collaboration with OQC and NVIDIA, has developed a full quantum processing unit (QPU)-to-GPU reference design using an Agilex 9 Direct RF FPGA. This design addresses the increasing density challenges in quantum computing by integrating analog-to-digital and digital-to-analog converters directly into the FPGA, eliminating external RF components. Farhad Shafai, Head of Business Solutions & Marketing at Altera, stated that the companies will showcase the system at Supercomputing 2026 in Chicago.

Agilex 9 Direct RF FPGA Enables Low-Latency QPU Control

The new reference design uses frequency multiplexing to share a single radio frequency port, reducing the need for additional RF hardware as quantum systems scale in complexity. This approach addresses the increasing density challenges inherent in building larger quantum computers by minimizing the footprint of control electronics. The integrated converters streamline signal processing and contribute to a more compact overall design.

NVQLink connects the FPGA to GPUs, which take on processing-heavy tasks such as calibration and machine learning-based readout classification, while the FPGA manages real-time qubit control and resonator readout. “Advances in AI are opening new ways to build and operate quantum computers,” said Sam Stanwyck, Director of Quantum Product, adding that “NVQLink gives quantum builders an open architecture to put those advances to work.” The reference design also includes CUDA functions that drive the FPGA logic for QPU system functions, which improves the connection between the two processing platforms. The system’s architecture runs real-time qubit control, resonator readout and spectroscopy within the FPGA itself, with synchronized scheduling across all functions.

This real-time processing capability is critical for achieving the low latency required for precise quantum control, OQC says. “Precise control and fast feedback are essential to unlocking the capabilities of our quantum processors,” explained Simon Phillips, CPO at OQC.

“Altera’s FPGA technology provides a programmable, low-latency connection to our QPU, bringing control and readout together while allowing us to refine calibration and feedback as our systems evolve.” The integration simplifies development and provides a scalable path for supporting an increasing number of qubits. Developers can also use a built-in emulation path within the reference design to test algorithms and validate the controller before deploying it with physical qubit hardware.

This feature accelerates the development process and allows for earlier identification and resolution of potential issues. “This demonstration shows how Altera’s Direct RF technology can anchor the control layer of quantum systems, and it gives developers a clear path to build on.” The full system reference design will be demonstrated at Supercomputing 2026 (SC26) in Chicago, and the Agilex 9 Direct RF quantum system controller reference design with NVQLink is expected to be released in January 2027.

Our work with OQC and NVIDIA uses FPGAs to demonstrate a full system reference design in a real-world setting, from the quantum processor through the FPGA to the GPU.

Farhad Shafai, Head of Business Solutions & Marketing at Altera
Stay current

See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.

Avatar of Ivy Delaney

Ivy Delaney

Ivy Delaney has been working with neural networks and machine learning since the mid-nineties, back when a couple of hidden layers and a long afternoon of training counted as ambitious. She has watched the field go from academic curiosity to the thing quietly running underneath everything, and she brings that long view to quantum computing. For Quantum Zeitgeist she covers the ground where the two fields meet. That means quantum machine learning and the variational algorithms it leans on, and it also means the less glamorous but more interesting story of classical machine learning already doing real work inside quantum machines, decoding error-correcting codes, calibrating noisy hardware and learning the error models that simulators depend on. She writes about the hardware those algorithms have to run on too, and about the post-quantum cryptography scramble that the same hardware has set off. Her stories typically start with the paper, whether that is peer-reviewed work, conference proceedings or an arXiv preprint, with the source linked so you can hold a claim up against the research it came from. She is unimpressed by benchmarks that will not say what they beat, and by demonstrations that only work in the press release.

Latest Posts by Ivy Delaney: