Supercomputing centres in Asia and Europe adopt NVQLink; Quantinuum Helios decodes errors in real time over it

NVIDIA/CUDA-Q and NVQLink

At SC25, NVIDIA said more than a dozen supercomputing centres in Asia, Europe and the Middle East are adopting NVQLink to attach quantum processors to its Grace Blackwell systems. Quantinuum showed the first real-time use of a scalable decoder for qLDPC error-correcting codes, with CUDA-Q running the decoding on NVIDIA GPUs.

• Adopters include RIKEN, AIST G-QuAT, KISTI, Jülich, GENCI, CINECA, the UK National Quantum Computing Centre and Pawsey.
• NVQLink specs: under 4 µs GPU-QPU latency, 400 Gb/s throughput and 40 PFLOPS of FP4 AI compute.
• Helios's decoder reacted in 67 µs, well inside the 2 ms the machine allows.