Code to implement the approximate Gibbs sampler for efficient inference of the hierarchical Bayesian model for grouped count data. The performance of the proposed sampler is compared against that of the start-of-the-art algorithm, the No-U-Turn-Sampler (NUTS) used by default in Stan.