The brms package in R implements Bayesian multilevel models using the Stan probabilistic programming language. It supports a wide range of distributions and link functions, allowing users to fit various types of models, including linear, robust linear, binomial, Poisson, survival, ordinal, zero-inflated, hurdle, and non-linear models. The package offers flexible prior specifications and allows for autocorrelation, user-defined covariance structures, censored data, and meta-analytic standard errors. Model fit can be assessed using the Watanabe-Akaike information criterion (WAIC) and leave-one-out cross-validation (LOO). The article provides an overview of the package, including its structure, examples, and comparisons with other R packages for multilevel modeling. It highlights the advantages of using Stan for Bayesian inference, such as the ability to incorporate prior knowledge and the efficiency of Hamiltonian Monte Carlo (HMC) and No-U-Turn Sampler (NUTS) algorithms. The package is available on CRAN and can be installed using R. The article also discusses the installation requirements, including a C++ compiler, and provides detailed instructions on how to fit models using the brm function. Examples are provided to illustrate the use of the package, including a model for recurrence times in kidney patients and a model for ordinal data from a crossover trial. The article concludes with a comparison of brms with other R packages for multilevel modeling, highlighting its strengths and flexibility.The brms package in R implements Bayesian multilevel models using the Stan probabilistic programming language. It supports a wide range of distributions and link functions, allowing users to fit various types of models, including linear, robust linear, binomial, Poisson, survival, ordinal, zero-inflated, hurdle, and non-linear models. The package offers flexible prior specifications and allows for autocorrelation, user-defined covariance structures, censored data, and meta-analytic standard errors. Model fit can be assessed using the Watanabe-Akaike information criterion (WAIC) and leave-one-out cross-validation (LOO). The article provides an overview of the package, including its structure, examples, and comparisons with other R packages for multilevel modeling. It highlights the advantages of using Stan for Bayesian inference, such as the ability to incorporate prior knowledge and the efficiency of Hamiltonian Monte Carlo (HMC) and No-U-Turn Sampler (NUTS) algorithms. The package is available on CRAN and can be installed using R. The article also discusses the installation requirements, including a C++ compiler, and provides detailed instructions on how to fit models using the brm function. Examples are provided to illustrate the use of the package, including a model for recurrence times in kidney patients and a model for ordinal data from a crossover trial. The article concludes with a comparison of brms with other R packages for multilevel modeling, highlighting its strengths and flexibility.