The paper introduces Ambiguity Clustering (AC), a new decoding algorithm for quantum low-density parity check (qLDPC) codes, which are more efficient and accurate than the state-of-the-art Belief Propagation followed by Ordered Statistics Decoding (BP-OSD). AC divides the measurement data into clusters that are decoded independently, significantly reducing computational complexity while maintaining or improving logical fidelity. The algorithm is benchmarked on bivariate bicycle codes, showing a speedup of 1 to 3 orders of magnitude compared to BP-OSD at realistic error rates. The CPU implementation of AC is fast enough to decode the 144-qubit Gross code in real time for neutral atom and trapped ion systems, making it suitable for practical quantum computing applications. The key innovation in AC is the dynamic selection of pivots during Gaussian elimination, which helps form a block structure in the parity check matrix, allowing for efficient cluster analysis. This approach turns a multiplicative cost into an additive one, significantly improving the efficiency of the decoding process.The paper introduces Ambiguity Clustering (AC), a new decoding algorithm for quantum low-density parity check (qLDPC) codes, which are more efficient and accurate than the state-of-the-art Belief Propagation followed by Ordered Statistics Decoding (BP-OSD). AC divides the measurement data into clusters that are decoded independently, significantly reducing computational complexity while maintaining or improving logical fidelity. The algorithm is benchmarked on bivariate bicycle codes, showing a speedup of 1 to 3 orders of magnitude compared to BP-OSD at realistic error rates. The CPU implementation of AC is fast enough to decode the 144-qubit Gross code in real time for neutral atom and trapped ion systems, making it suitable for practical quantum computing applications. The key innovation in AC is the dynamic selection of pivots during Gaussian elimination, which helps form a block structure in the parity check matrix, allowing for efficient cluster analysis. This approach turns a multiplicative cost into an additive one, significantly improving the efficiency of the decoding process.