Pass@1 is a critical evaluation metric for mathematical and logical reasoning tasks in the domains of large language models and reasoning. It measures the proportion of correct answers in a single sampling generation, reflecting the model's reliable accuracy in deterministic reasoning scenarios, and is closer to the usage habit of single-response interactions in actual deployments compared to Pass@k. It is typically reported as a percentage or a score between 0 and 1. This metric is widely used in relevant benchmarks and papers, effectively reflecting model performance in this dimension. Higher values are better.