diff --git a/DESCRIPTION b/DESCRIPTION
index 3d55cb48..7ce3e65e 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -29,7 +29,7 @@ URL: https://mc-stan.org/cmdstanr/, https://discourse.mc-stan.org
 BugReports: https://github.com/stan-dev/cmdstanr/issues
 Encoding: UTF-8
 LazyData: true
-RoxygenNote: 7.3.0
+RoxygenNote: 7.3.1
 Roxygen: list(markdown = TRUE, r6 = FALSE)
 SystemRequirements: CmdStan (https://mc-stan.org/users/interfaces/cmdstan)
 Depends:
diff --git a/R/cmdstanr-package.R b/R/cmdstanr-package.R
index 305241bf..ce6bc525 100644
--- a/R/cmdstanr-package.R
+++ b/R/cmdstanr-package.R
@@ -26,6 +26,9 @@
 #'   [_Getting started with CmdStanR_](https://mc-stan.org/cmdstanr/articles/cmdstanr.html)
 #'   demonstrates the basic functionality of the package.
 #'
+#'   For a list of global [options][base::options()] see
+#'   [cmdstanr_global_options].
+#'
 #' @template seealso-docs
 #' @inherit cmdstan_model examples
 #' @import R6
diff --git a/R/generics.R b/R/generics.R
index 663cb859..656cf9a1 100644
--- a/R/generics.R
+++ b/R/generics.R
@@ -1,9 +1,8 @@
-
 #' Coercion methods for CmdStan objects
 #'
-#' These methods are used to coerce objects into `cmdstanr` objects.
-#' Primarily intended for other packages to use when interfacing
-#' with `cmdstanr`.
+#' These are generic functions intended to primarily be used by developers of
+#' packages that interface with on CmdStanR. Developers can define methods on
+#' top of these generics to coerce objects into CmdStanR's fitted model objects.
 #'
 #' @param object to be coerced
 #' @param ... additional arguments
diff --git a/R/model.R b/R/model.R
index 289558f0..0b4ba47e 100644
--- a/R/model.R
+++ b/R/model.R
@@ -1123,7 +1123,7 @@ sample <- function(data = NULL,
                    refresh = NULL,
                    init = NULL,
                    save_latent_dynamics = FALSE,
-                   output_dir = NULL,
+                   output_dir = getOption("cmdstanr_output_dir"),
                    output_basename = NULL,
                    sig_figs = NULL,
                    chains = 4,
@@ -1334,7 +1334,7 @@ sample_mpi <- function(data = NULL,
                        refresh = NULL,
                        init = NULL,
                        save_latent_dynamics = FALSE,
-                       output_dir = NULL,
+                       output_dir = getOption("cmdstanr_output_dir"),
                        output_basename = NULL,
                        chains = 1,
                        chain_ids = seq_len(chains),
@@ -1484,7 +1484,7 @@ optimize <- function(data = NULL,
                      refresh = NULL,
                      init = NULL,
                      save_latent_dynamics = FALSE,
-                     output_dir = NULL,
+                     output_dir = getOption("cmdstanr_output_dir"),
                      output_basename = NULL,
                      sig_figs = NULL,
                      threads = NULL,
@@ -1622,7 +1622,7 @@ laplace <- function(data = NULL,
                     refresh = NULL,
                     init = NULL,
                     save_latent_dynamics = FALSE,
-                    output_dir = NULL,
+                    output_dir = getOption("cmdstanr_output_dir"),
                     output_basename = NULL,
                     sig_figs = NULL,
                     threads = NULL,
@@ -1769,7 +1769,7 @@ variational <- function(data = NULL,
                         refresh = NULL,
                         init = NULL,
                         save_latent_dynamics = FALSE,
-                        output_dir = NULL,
+                        output_dir = getOption("cmdstanr_output_dir"),
                         output_basename = NULL,
                         sig_figs = NULL,
                         threads = NULL,
@@ -1908,7 +1908,7 @@ pathfinder <- function(data = NULL,
                        refresh = NULL,
                        init = NULL,
                        save_latent_dynamics = FALSE,
-                       output_dir = NULL,
+                       output_dir = getOption("cmdstanr_output_dir"),
                        output_basename = NULL,
                        sig_figs = NULL,
                        opencl_ids = NULL,
@@ -2060,7 +2060,7 @@ CmdStanModel$set("public", name = "pathfinder", value = pathfinder)
 generate_quantities <- function(fitted_params,
                                 data = NULL,
                                 seed = NULL,
-                                output_dir = NULL,
+                                output_dir = getOption("cmdstanr_output_dir"),
                                 output_basename = NULL,
                                 sig_figs = NULL,
                                 parallel_chains = getOption("mc.cores", 1),
@@ -2125,7 +2125,7 @@ CmdStanModel$set("public", name = "generate_quantities", value = generate_quanti
 diagnose <- function(data = NULL,
                      seed = NULL,
                      init = NULL,
-                     output_dir = NULL,
+                     output_dir = getOption("cmdstanr_output_dir"),
                      output_basename = NULL,
                      epsilon = NULL,
                      error = NULL) {
diff --git a/R/options.R b/R/options.R
new file mode 100644
index 00000000..7948908a
--- /dev/null
+++ b/R/options.R
@@ -0,0 +1,46 @@
+#' CmdStanR global options
+#'
+#' These options can be set via [options()] for an entire \R session.
+#'
+#' @name cmdstanr_global_options
+#'
+#' @details
+#'
+#' * `cmdstanr_draws_format`: Which format provided by the \pkg{posterior}
+#' package should be used when returning the posterior or approximate posterior
+#' draws? The default depends on the model fitting method. See
+#' [draws][fit-method-draws] for more details.
+#'
+#' * `cmdstanr_force_recompile`: Should the default be to recompile models
+#' even if there were no Stan code changes since last compiled?  See
+#' [compile][model-method-compile] for more details. The default is `FALSE`.
+#'
+#' * `cmdstanr_max_rows`: The maximum number of rows of output to print when
+#' using the [`$print()`][fit-method-summary] method. The default is 10.
+#'
+#' * `cmdstanr_no_ver_check`: Should the check for a more recent version of
+#' CmdStan be disabled? The default is `FALSE`.
+#'
+#' * `cmdstanr_output_dir`: The directory where CmdStan should write its output
+#' CSV files when fitting models. The default is a temporary directory. Files in
+#' a temporary directory are removed as part of \R garbage collection, while
+#' files in an explicitly defined directory are not automatically deleted.
+#'
+#' * `cmdstanr_verbose`: Should more information be printed
+#' when compiling or running models, including showing how CmdStan was called
+#' internally? The default is `FALSE`.
+#'
+#' * `cmdstanr_warn_inits`: Should a warning be thrown if initial values are
+#' only provided for a subset of parameters? The default is `TRUE`.
+#'
+#' * `cmdstanr_write_stan_file_dir`: The directory where [write_stan_file()]
+#' should write Stan files. The default is a temporary directory. Files in
+#' a temporary directory are removed as part of \R garbage collection, while
+#' files in an explicitly defined directory are not automatically deleted.
+#'
+#' * `mc.cores`: The number of cores to use for various parallelization tasks
+#' (e.g. running MCMC chains, installing CmdStan). The default depends on the
+#' use case and is documented with the methods that make use of `mc.cores`.
+#'
+#'
+NULL
diff --git a/R/zzz.R b/R/zzz.R
index 0a255258..860eb894 100644
--- a/R/zzz.R
+++ b/R/zzz.R
@@ -10,9 +10,16 @@ startup_messages <- function() {
   }
 
   skip_version_check <- isTRUE(getOption(
-    "CMDSTANR_NO_VER_CHECK",
-    default = identical(tolower(Sys.getenv("CMDSTANR_NO_VER_CHECK")), "true")
+    "cmdstanr_no_ver_check",
+    default = identical(tolower(Sys.getenv("cmdstanr_no_ver_check")), "true")
   ))
+  if (!skip_version_check) {
+    # check if they used the old all caps version
+    skip_version_check <- isTRUE(getOption(
+      "CMDSTANR_NO_VER_CHECK",
+      default = identical(tolower(Sys.getenv("CMDSTANR_NO_VER_CHECK")), "true")
+    ))
+  }
   if (!skip_version_check) {
     latest_version <- try(suppressWarnings(latest_released_version(retries = 0)), silent = TRUE)
     current_version <- try(cmdstan_version(), silent = TRUE)
@@ -21,7 +28,7 @@ startup_messages <- function() {
         && latest_version > current_version) {
       packageStartupMessage(
         "\nA newer version of CmdStan is available. See ?install_cmdstan() to install it.",
-        "\nTo disable this check set option or environment variable CMDSTANR_NO_VER_CHECK=TRUE."
+        "\nTo disable this check set option or environment variable cmdstanr_no_ver_check=TRUE."
       )
     }
   }
diff --git a/man-roxygen/model-common-args.R b/man-roxygen/model-common-args.R
index ca609ebe..f65eb21c 100644
--- a/man-roxygen/model-common-args.R
+++ b/man-roxygen/model-common-args.R
@@ -60,7 +60,8 @@
 #'   methods there will be a single file. For interactive use this can typically
 #'   be left at `NULL` (temporary directory) since CmdStanR makes the CmdStan
 #'   output (posterior draws and diagnostics) available in \R via methods of the
-#'   fitted model objects. The behavior of `output_dir` is as follows:
+#'   fitted model objects. This can be set for an entire \R session using
+#'   `options(cmdstanr_output_dir)`. The behavior of `output_dir` is as follows:
 #'   * If `NULL` (the default), then the CSV files are written to a temporary
 #'   directory and only saved permanently if the user calls one of the `$save_*`
 #'   methods of the fitted model object (e.g.,
diff --git a/man/cmdstan_coercion.Rd b/man/cmdstan_coercion.Rd
index d2d2b483..33a36d3a 100644
--- a/man/cmdstan_coercion.Rd
+++ b/man/cmdstan_coercion.Rd
@@ -31,7 +31,7 @@ as.CmdStanDiagnose(object, ...)
 \item{...}{additional arguments}
 }
 \description{
-These methods are used to coerce objects into \code{cmdstanr} objects.
-Primarily intended for other packages to use when interfacing
-with \code{cmdstanr}.
+These are generic functions intended to primarily be used by developers of
+packages that interface with on CmdStanR. Developers can define methods on
+top of these generics to coerce objects into CmdStanR's fitted model objects.
 }
diff --git a/man/cmdstanr-package.Rd b/man/cmdstanr-package.Rd
index c98b0b3a..87bedc0d 100644
--- a/man/cmdstanr-package.Rd
+++ b/man/cmdstanr-package.Rd
@@ -65,6 +65,9 @@ you already have CmdStan installed see \code{\link[=cmdstan_model]{cmdstan_model
 otherwise see \code{\link[=install_cmdstan]{install_cmdstan()}} to install CmdStan. The vignette
 \href{https://mc-stan.org/cmdstanr/articles/cmdstanr.html}{\emph{Getting started with CmdStanR}}
 demonstrates the basic functionality of the package.
+
+For a list of global \link[base:options]{options} see
+\link{cmdstanr_global_options}.
 }
 
 \examples{
diff --git a/man/cmdstanr_global_options.Rd b/man/cmdstanr_global_options.Rd
new file mode 100644
index 00000000..a011b81a
--- /dev/null
+++ b/man/cmdstanr_global_options.Rd
@@ -0,0 +1,39 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/options.R
+\name{cmdstanr_global_options}
+\alias{cmdstanr_global_options}
+\title{CmdStanR global options}
+\description{
+These options can be set via \code{\link[=options]{options()}} for an entire \R session.
+}
+\details{
+\itemize{
+\item \code{cmdstanr_draws_format}: Which format provided by the \pkg{posterior}
+package should be used when returning the posterior or approximate posterior
+draws? The default depends on the model fitting method. See
+\link[=fit-method-draws]{draws} for more details.
+\item \code{cmdstanr_force_recompile}: Should the default be to recompile models
+even if there were no Stan code changes since last compiled?  See
+\link[=model-method-compile]{compile} for more details. The default is \code{FALSE}.
+\item \code{cmdstanr_max_rows}: The maximum number of rows of output to print when
+using the \code{\link[=fit-method-summary]{$print()}} method. The default is 10.
+\item \code{cmdstanr_no_ver_check}: Should the check for a more recent version of
+CmdStan be disabled? The default is \code{FALSE}.
+\item \code{cmdstanr_output_dir}: The directory where CmdStan should write its output
+CSV files when fitting models. The default is a temporary directory. Files in
+a temporary directory are removed as part of \R garbage collection, while
+files in an explicitly defined directory are not automatically deleted.
+\item \code{cmdstanr_verbose}: Should more information be printed
+when compiling or running models, including showing how CmdStan was called
+internally? The default is \code{FALSE}.
+\item \code{cmdstanr_warn_inits}: Should a warning be thrown if initial values are
+only provided for a subset of parameters? The default is \code{TRUE}.
+\item \code{cmdstanr_write_stan_file_dir}: The directory where \code{\link[=write_stan_file]{write_stan_file()}}
+should write Stan files. The default is a temporary directory. Files in
+a temporary directory are removed as part of \R garbage collection, while
+files in an explicitly defined directory are not automatically deleted.
+\item \code{mc.cores}: The number of cores to use for various parallelization tasks
+(e.g. running MCMC chains, installing CmdStan). The default depends on the
+use case and is documented with the methods that make use of \code{mc.cores}.
+}
+}
diff --git a/man/model-method-check_syntax.Rd b/man/model-method-check_syntax.Rd
index a646a5e1..68366fb5 100644
--- a/man/model-method-check_syntax.Rd
+++ b/man/model-method-check_syntax.Rd
@@ -86,8 +86,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-compile.Rd b/man/model-method-compile.Rd
index d295eedc..7bfa47d7 100644
--- a/man/model-method-compile.Rd
+++ b/man/model-method-compile.Rd
@@ -157,8 +157,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-diagnose.Rd b/man/model-method-diagnose.Rd
index 9a0acd31..631ce6f6 100644
--- a/man/model-method-diagnose.Rd
+++ b/man/model-method-diagnose.Rd
@@ -9,7 +9,7 @@ diagnose(
   data = NULL,
   seed = NULL,
   init = NULL,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   epsilon = NULL,
   error = NULL
@@ -68,7 +68,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
@@ -129,8 +130,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-expose_functions.Rd b/man/model-method-expose_functions.Rd
index a62f7bb8..b7d42231 100644
--- a/man/model-method-expose_functions.Rd
+++ b/man/model-method-expose_functions.Rd
@@ -77,8 +77,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-format.Rd b/man/model-method-format.Rd
index 2aa34f18..d24010a4 100644
--- a/man/model-method-format.Rd
+++ b/man/model-method-format.Rd
@@ -106,8 +106,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-generate-quantities.Rd b/man/model-method-generate-quantities.Rd
index bf25602e..21ce47a2 100644
--- a/man/model-method-generate-quantities.Rd
+++ b/man/model-method-generate-quantities.Rd
@@ -9,7 +9,7 @@ generate_quantities(
   fitted_params,
   data = NULL,
   seed = NULL,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   sig_figs = NULL,
   parallel_chains = getOption("mc.cores", 1),
@@ -62,7 +62,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
@@ -178,8 +179,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-laplace.Rd b/man/model-method-laplace.Rd
index b033fbe3..2ebb00b3 100644
--- a/man/model-method-laplace.Rd
+++ b/man/model-method-laplace.Rd
@@ -11,7 +11,7 @@ laplace(
   refresh = NULL,
   init = NULL,
   save_latent_dynamics = FALSE,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   sig_figs = NULL,
   threads = NULL,
@@ -83,7 +83,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
@@ -214,8 +215,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-generate-quantities}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-optimize.Rd b/man/model-method-optimize.Rd
index dcf77444..dbed1150 100644
--- a/man/model-method-optimize.Rd
+++ b/man/model-method-optimize.Rd
@@ -11,7 +11,7 @@ optimize(
   refresh = NULL,
   init = NULL,
   save_latent_dynamics = FALSE,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   sig_figs = NULL,
   threads = NULL,
@@ -97,7 +97,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
@@ -332,8 +333,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-generate-quantities}},
 \code{\link{model-method-laplace}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-pathfinder.Rd b/man/model-method-pathfinder.Rd
index 85fc9236..d16ad170 100644
--- a/man/model-method-pathfinder.Rd
+++ b/man/model-method-pathfinder.Rd
@@ -11,7 +11,7 @@ pathfinder(
   refresh = NULL,
   init = NULL,
   save_latent_dynamics = FALSE,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   sig_figs = NULL,
   opencl_ids = NULL,
@@ -102,7 +102,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
@@ -357,8 +358,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-generate-quantities}},
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}},
 \code{\link{model-method-variational}}
 }
diff --git a/man/model-method-sample.Rd b/man/model-method-sample.Rd
index 526c9c88..51b4ce45 100644
--- a/man/model-method-sample.Rd
+++ b/man/model-method-sample.Rd
@@ -11,7 +11,7 @@ sample(
   refresh = NULL,
   init = NULL,
   save_latent_dynamics = FALSE,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   sig_figs = NULL,
   chains = 4,
@@ -115,7 +115,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
diff --git a/man/model-method-sample_mpi.Rd b/man/model-method-sample_mpi.Rd
index ca7203da..078461bb 100644
--- a/man/model-method-sample_mpi.Rd
+++ b/man/model-method-sample_mpi.Rd
@@ -13,7 +13,7 @@ sample_mpi(
   refresh = NULL,
   init = NULL,
   save_latent_dynamics = FALSE,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   chains = 1,
   chain_ids = seq_len(chains),
@@ -114,7 +114,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
diff --git a/man/model-method-variables.Rd b/man/model-method-variables.Rd
index dc80ed9a..87e9d73e 100644
--- a/man/model-method-variables.Rd
+++ b/man/model-method-variables.Rd
@@ -46,8 +46,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variational}}
 }
 \concept{CmdStanModel methods}
diff --git a/man/model-method-variational.Rd b/man/model-method-variational.Rd
index 3678f11e..87decfd9 100644
--- a/man/model-method-variational.Rd
+++ b/man/model-method-variational.Rd
@@ -11,7 +11,7 @@ variational(
   refresh = NULL,
   init = NULL,
   save_latent_dynamics = FALSE,
-  output_dir = NULL,
+  output_dir = getOption("cmdstanr_output_dir"),
   output_basename = NULL,
   sig_figs = NULL,
   threads = NULL,
@@ -98,7 +98,8 @@ its output CSV files. For MCMC there will be one file per chain; for other
 methods there will be a single file. For interactive use this can typically
 be left at \code{NULL} (temporary directory) since CmdStanR makes the CmdStan
 output (posterior draws and diagnostics) available in \R via methods of the
-fitted model objects. The behavior of \code{output_dir} is as follows:
+fitted model objects. This can be set for an entire \R session using
+\code{options(cmdstanr_output_dir)}. The behavior of \code{output_dir} is as follows:
 \itemize{
 \item If \code{NULL} (the default), then the CSV files are written to a temporary
 directory and only saved permanently if the user calls one of the \verb{$save_*}
@@ -333,8 +334,8 @@ Other CmdStanModel methods:
 \code{\link{model-method-laplace}},
 \code{\link{model-method-optimize}},
 \code{\link{model-method-pathfinder}},
-\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-sample}},
+\code{\link{model-method-sample_mpi}},
 \code{\link{model-method-variables}}
 }
 \concept{CmdStanModel methods}