Self-Consistency Score is a key evaluation metric for mathematical and logical reasoning tasks in the field of large language models and inference. It measures the proportion of consistent conclusions provided by the model across multiple responses or equivalent formulations, reflecting the internal self-consistency of logical reasoning results. A higher score indicates a more robust reasoning process. This metric is typically reported as a percentage or a score between 0 and 1. It is widely used in relevant benchmarks and papers, effectively reflecting model performance in this dimension. Higher values are better. It helps researchers and engineers quantify actual performance differences of models in mathematical and logical reasoning scenarios, facilitating horizontal comparison of different methods or model versions.