Hardware
Quantum X Labs reports up to 20% fewer decoding failures on IBM processors
Quantum X Labs says preliminary benchmarks on IBM quantum processors showed up to 20% fewer logical decoding failures. The company expanded its AI-powered benchmarking program to IBM Fez, IBM Kingston and IBM Pittsburgh. Its workflow combines experimental quantum measurements with GPU-accelerated AI training and uses noise-model calibration to improve how decoders identify and correct errors. The company says the approach builds on earlier work on Google quantum hardware, and the results are described as preliminary; the article does not say they have been independently verified or peer reviewed.
Why it mattersQuantum processors are susceptible to environmental interference and hardware imperfections, so better decoding is one route to more reliable error correction. Quantum X Labs says it expects efficient error correction to become essential as processors grow in size and complexity, though these are company-reported, preliminary figures.