SANTA CLARA — Quantum computing is poised to revolutionize industries such as drug discovery and logistics, yet a significant obstacle remains: noise. Environmental interactions and hardware imperfections introduce disturbances that limit today’s quantum devices to just hundreds of operations before computations degrade. Addressing this challenge requires logical qubits, which combine tens or hundreds of physical qubits to correct noise-induced errors and maintain computational integrity.
Infleqtion, a leading quantum hardware developer, has taken a significant step toward this goal by leveraging NVIDIA’s CUDA-Q platform to design and execute an experiment involving two logical qubits. This experiment marked the first-ever demonstration of a materials science quantum algorithm, specifically the single-impurity Anderson model, using logical qubits—a critical milestone for advancing quantum materials science applications.
Logical qubits are notoriously difficult to create, requiring highly accurate modeling and error correction. Infleqtion achieved this breakthrough by utilizing CUDA-Q’s GPU-accelerated simulation tools to model its quantum system with precision. The experiment was developed and tested entirely within CUDA-Q’s simulators, allowing Infleqtion to transition seamlessly to its physical hardware, the Sqale neutral atom quantum processor, with minimal adjustments.
This achievement underscores the importance of integrating advanced quantum hardware with AI-powered supercomputers. As quantum computing evolves toward large-scale, error-corrected systems, collaboration between quantum platforms and AI-driven technologies will be essential for overcoming scalability challenges.
NVIDIA continues to partner with innovators like Infleqtion to drive forward the research necessary to develop accelerated quantum supercomputers, paving the way for quantum computing to unlock transformative applications across industries.
Read the NVIDIA Technical Blog on this news and learn more about NVIDIA’s quantum computing platforms.