Quantum X Labs Inc. (NASDAQ: QXL) said its AI-powered quantum error-correction decoder achieved improved performance against Google's published benchmarks in testing using Google's public surface-code dataset from a quantum-hardware experiment.
The decoder, which combines quantum-code structure, syndrome information, and AI-based error weighting, was trained exclusively on synthetic samples and did not use real hardware data from the Google dataset. In the tests, it outperformed Google's correlated-matching and PyMatching benchmark results for the same configuration, according to the company.
Prof. Nir Sharon, Chief Quantum Technology Scientist at Quantum X Labs, said the results mark progress toward evaluating AI-driven quantum error correction against real hardware behavior rather than simulations. 'Our updated decoder improved performance against matching-family benchmarks in this experiment while training only on synthetic data,' he stated.
Quantum X Labs, headquartered in Tel Aviv, and its subsidiary Quantum X Labs Ltd. focus on quantum technology, digital advertising, computing, and enterprise AI solutions. The subsidiary develops quantum algorithms for transportation, drug discovery, and security applications.
The company plans to replicate and extend the results across additional device centers and code configurations, with next steps including evaluation on real hardware and integration with NVIDIA accelerated computing and NVIDIA CUDA-Q workflows. Quantum X Labs also aims to conduct IQCC syndrome experiments as part of its roadmap.
The announcement was published on August 21, 2026.
