Quantum X Labs Outperforms Baselines with AI-Driven Quantum Error Correction Decoder
Quantum software developer Quantum X Labs Inc. (Nasdaq: QXL) has announced improved performance results from its AI-driven quantum error correction decoder program. The company tested its updated model on Google's public surface-code experimental dataset, achieving higher decoding accuracy compared to standard matching-family baselines.
The AI-QEC Decoder Architecture uses a synthetic-to-real generalization pipeline and leverages NVIDIA CUDA-Q and GPU acceleration. Despite being trained exclusively on synthetic samples, the model outperformed PyMatching and correlated-matching benchmark results for the same surface-code configuration.
Quantum X Labs' updated decoder model demonstrates its ability to generalize to physical device noise without requiring extensive retraining on real hardware shots. The company plans to extend and replicate this pipeline across additional physical hardware backends, code topologies, and device centers.