IonQ Plans Superion 256 Installation at NVIDIA Quantum Center
IonQ announced it will install a Superion 256 quantum computer at NVIDIA’s Accelerated Quantum Research Center. This marks the first time IonQ’s trapped-ion hardware will be directly integrated into NVIDIA’s GPU-accelerated research stack, rather than accessed via a cloud queue, making it the first on-premise quantum system at the facility. The company announced this plan on September 23, 2026, with installation slated for 2027.
IonQ’s Contribution to the Center
According to a press release, the Superion 256 will connect to an NVIDIA GB200 NVL72 system through NVQLink, with job coordination managed by the open-source CUDA-Q platform. Meanwhile, IonQ introduced its sixth-generation Superion line on September 8, 2026, and states that the 256 system is available for order now, with initial customer deliveries anticipated next year. Per NVIDIA’s description of the center, the goal of NVAQC is to address the challenge of scaling quantum processors with classical supercomputers.
Niccolo de Masi, IonQ Chairman and CEO, described the deployment as “the beginning of an exciting era of hybrid quantum-classical computing.” He also highlighted the company’s roadmap towards 10,000 qubits and full fault tolerance through its SkyWater CMOS foundry. These performance and manufacturability claims originate from IonQ. Timothy Costa, head of quantum at NVIDIA, noted that the Superion 256 will facilitate broader ecosystem integration of QPUs with AI infrastructure using open-source software.
Research Program Details and Unanswered Questions
The collaborative research will focus on quantum-GPU system design and hybrid software. Target applications include portfolio optimization and risk modeling for finance, materials science, and computational chemistry for drug discovery. The partners intend to publish open results and a co-design guide. Earlier in September, at IEEE Quantum Week, IonQ, NVIDIA, Oak Ridge National Laboratory, and the University of Tennessee published a study on pairing generative AI with distributed quantum algorithms for combinatorial optimization, executed on CUDA-Q.
This AQRC deployment aligns with IonQ’s strategy of partner-heavy, access-focused initiatives. The company is a founding partner in the George Mason-led Virginia Quantum Hub, and its NVQLink efforts extend to other regions, including a hybrid quantum-HPC alliance with KISTI in South Korea. The growing open-source ecosystem activity, as seen in Fujitsu’s open-sourced quantum application package, reflects a similar approach.