Reproduce “Engineering Low-Latency, High-Throughput Quantum Decoders on NVIDIA GPUs” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Stefano Markidis, Luca Pennati, Ivy Peng
The authors implement a GPU quantum decoder that runs iterative Relay belief propagation, retires samples once their candidate corrections match their syndromes, and sends unresolved samples to an OSD-0 fallback. Concurrent streams process independent batches. On simulated BB90 syndromes at error probability 0.003, RTX 5080 reaches 91% of GH200’s highest measured burst throughput with 8.9-times shorter median response; this is a benchmark, not a live-system demonstration.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.