Quantum X Labs Tests AI-Based Error Correction Decoder With NVIDIA CUDA-Q Tools

Insider Brief

• Quantum X Labs reported progress on its AI-driven quantum error-correction program using NVIDIA accelerated computing and CUDA-Q QEC software libraries.
• The company benchmarked its transformer-based QECCT decoder against the classical MWPM decoder in simulated toric-code and surface-code environments.
• QXL plans to extend testing toward experimental syndrome data from superconducting quantum hardware through its collaboration with IQCC.

Press release – Quantum X Labs Inc. (Nasdaq: QXL) (“Quantum X” or the “Company”), an advanced technologies company, today announced progress on its technical roadmap for its AI-driven quantum error-correction program.

The work is focused on reviewing QXL’s milestone results and draws on NVIDIA accelerated computing, the NVIDIA CUDA-Q QEC software libraries and, as QXL progresses from simulation-based validation toward hardware-derived syndrome data and future real-time decoding.

QXL has completed two meaningful development steps. First, the Company executed its Deep Quantum Error Correction (DQEC) workflow on an NVIDIA GPU in an AWS environment and benchmarked its transformer-based QECCT decoder against the classical Minimum-Weight Perfect Matching (MWPM) decoder across controlled toric-code noise configurations. QECCT outperformed MWPM in selected simulated regimes.

QXL also tested synthetic surface-code configurations modeled on Google’s public surface-code geometry and experiment structure, spanning multiple code distances. Across these scenarios, the QECCT decoder showed stable logical and bit error rates under varying physical error conditions.

These results are intended as an initial step in validating the approach within controlled simulation environments. Future work is expected to focus on extending these evaluations to publicly available experimental datasets and continuing to refine data pipelines and decoder workflows compatible with CUDA-Q QEC frameworks.

The broader roadmap also includes QXL’s planned work with IQCC, a Quantum Machines company, to generate hardware-derived syndrome data on superconducting quantum processing hardware. QXL is also reviewing where AI-based pre-decoder workflows using NVIDIA Ising, and low-latency optimization can add the greatest value across these stages of QXL’s roadmap.

“These results are meaningful because they move our program from cloud deployment into measured decoder performance and a surface-code data pipeline,” said Prof. Nir Sharon, Chief Quantum Technology Scientist at Quantum X Labs. “We are accelerating QXL’s roadmap with NVIDIA accelerated computing, CUDA-Q QEC and low-latency QEC expertise, while IQCC provides the path to hardware-derived syndrome data. This staged approach is designed to move us from offline benchmarking toward practical real-time QEC.”

QXL’s DQEC technology is based on a proprietary transformer architecture that uses QEC code structure and syndrome information to predict logical corrections. The program is intended to support multiple stabilizer-code workflows and to evaluate the decoder as a full decoder, pre-decoder or hybrid component within accelerated QEC systems.