diff --git a/NAMESPACE b/NAMESPACE
index 98da81f5..73546e6a 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -410,6 +410,7 @@ export(dhurdle_negbinomial)
 export(dhurdle_poisson)
 export(dinv_gaussian)
 export(dirichlet)
+export(dirichlet_multinomial)
 export(dlogistic_normal)
 export(dmulti_normal)
 export(dmulti_student_t)
diff --git a/man/brmsfamily.Rd b/man/brmsfamily.Rd
index fbf81457..ae29eae4 100644
--- a/man/brmsfamily.Rd
+++ b/man/brmsfamily.Rd
@@ -36,6 +36,7 @@
 \alias{zero_inflated_beta_binomial}
 \alias{categorical}
 \alias{multinomial}
+\alias{dirichlet_multinomial}
 \alias{cumulative}
 \alias{sratio}
 \alias{cratio}
@@ -153,6 +154,8 @@ categorical(link = "logit", refcat = NULL)
 
 multinomial(link = "logit", refcat = NULL)
 
+dirichlet_multinomial(link = "logit", link_phi = "log", refcat = NULL)
+
 cumulative(link = "logit", link_disc = "log", threshold = "flexible")
 
 sratio(link = "logit", link_disc = "log", threshold = "flexible")
@@ -171,7 +174,7 @@ supported: \code{gaussian}, \code{student}, \code{binomial},
 \code{inverse.gaussian}, \code{exponential}, \code{weibull},
 \code{frechet}, \code{Beta}, \code{dirichlet}, \code{von_mises},
 \code{asym_laplace}, \code{gen_extreme_value}, \code{categorical},
-\code{multinomial}, \code{cumulative}, \code{cratio}, \code{sratio},
+\code{multinomial}, \code{dirichlet_multinomial}, \code{cumulative}, \code{cratio}, \code{sratio},
 \code{acat}, \code{hurdle_poisson}, \code{hurdle_negbinomial},
 \code{hurdle_gamma}, \code{hurdle_lognormal}, \code{hurdle_cumulative},
 \code{zero_inflated_binomial}, \code{zero_inflated_beta_binomial},
@@ -224,8 +227,9 @@ consecutive thresholds to the same value, and
 \code{"sum_to_zero"} ensures the thresholds sum to zero.}
 
 \item{refcat}{Optional name of the reference response category used in
-\code{categorical}, \code{multinomial}, \code{dirichlet} and
-\code{logistic_normal} models. If \code{NULL} (the default), the first
+\code{categorical}, \code{multinomial}, \code{dirichlet},
+\code{dirichlet_multinomial} and \code{logistic_normal} models.
+If \code{NULL} (the default), the first
 category is used as the reference. If \code{NA}, all categories will be
 predicted, which requires strong priors or carefully specified predictor
 terms in order to lead to an identified model.}
@@ -259,8 +263,9 @@ Below, we list common use cases for the different families.
   can be used for binary regression (i.e., most commonly logistic
   regression).}
 
-  \item{Families \code{categorical} and \code{multinomial} can be used for
-  multi-logistic regression when there are more than two possible outcomes.}
+  \item{Families \code{categorical}, \code{multinomial} and
+  \code{dirichlet_multinomial} can be used for multi-logistic regression
+  when there are more than two possible outcomes.}
 
   \item{Families \code{cumulative}, \code{cratio} ('continuation ratio'),
   \code{sratio} ('stopping ratio'), and \code{acat} ('adjacent category')
@@ -333,8 +338,8 @@ Below, we list common use cases for the different families.
   \code{acat}, and \code{hurdle_cumulative} support \code{logit},
   \code{probit}, \code{probit_approx}, \code{cloglog}, and \code{cauchit}.}
 
-  \item{Families \code{categorical}, \code{multinomial}, and \code{dirichlet}
-  support \code{logit}.}
+  \item{Families \code{categorical}, \code{multinomial},
+  \code{dirichlet_multinomiak}and \code{dirichlet} support \code{logit}.}
 
   \item{Families \code{Gamma}, \code{weibull}, \code{exponential},
   \code{frechet}, and \code{hurdle_gamma} support