NVIDIA and IBM link quantum processors to GPUs for error-correction gains
NVIDIA, IBM and Fermilab are advancing a modular quantum-computing co-design platform to connect quantum processors with accelerated computing and support research into error correction and applications in high-energy and nuclear physics. NVIDIA and IBM also detailed expanded work linking IBM quantum computers with NVIDIA DGX systems, including a CUDA-Q decoding algorithm reported to cut a quantum error-correction computation from 12 weeks to 45 minutes. Separately, NVIDIA’s Slurm updates are intended to let quantum processors be scheduled alongside CPUs and GPUs, with IQM systems among the early integrations aimed at bringing hybrid quantum-classical jobs into standard supercomputing workflows. Altera, Oxford Quantum Circuits and NVIDIA are developing an FPGA-based controller that links an OQC quantum processor to NVIDIA GPUs, combining real-time qubit control and readout with accelerated computing while aiming to reduce the size, power use and latency of control systems. The efforts reflect a broader push to make quantum hardware easier to integrate into existing supercomputing infrastructure and scale for scientific research.






