diff --git a/NAMESPACE b/NAMESPACE
index b5a60b3c..4e2af418 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -7,6 +7,7 @@ S3method(tp,data.frame)
 export(DGP)
 export(Evaluator)
 export(Experiment)
+export(HTML)
 export(Method)
 export(Visualizer)
 export(add_dgp)
diff --git a/R/docs.R b/R/docs.R
index 18b8a587..92794983 100644
--- a/R/docs.R
+++ b/R/docs.R
@@ -167,6 +167,17 @@ create_doc_template <- function(experiment, save_dir) {
 #' @param viz_order Vector of `Visualizer` names in their desired order for
 #'   display. By default, the report will display the `Visualizer` results
 #'   in the order that they were computed.
+#' @param eval_cache File extension of the cached evaluator results to read in.
+#'   Typically `".rds"` or `"none"`, but can be any file extension where
+#'   evaluator results are stored as `eval_results.ext` and can be read in
+#'   using `data.table::fread(eval_results.ext)`. If `"none"`, evaluator results
+#'   are computed using the experiment via [evaluate_experiment()].
+#' @param viz_cache File extension of the cached visualizer results to read in.
+#'   Typically `".rds"` or `"none"`, but can be any (image) file extension
+#'   (e.g., `"png"`, `"jpg"`) where the visualizer results have been previously
+#'   stored as separate `{visualizer_name}.ext` images (typically using
+#'   [export_visualizers()]). If `"none"`, visualizer results are computed using
+#'   the experiment via [visualize_experiment()].
 #' @param use_icons Logical indicating whether or not to use fontawesome icons.
 #' @param verbose Level of verboseness (0, 1, 2) when knitting R Markdown.
 #'   Default is 2.
@@ -189,7 +200,9 @@ render_docs <- function(experiment, save_dir, write_rmd = FALSE,
                         output_file = NULL, output_format = vthemes::vmodern(),
                         title = NULL, author = "",
                         show_code = TRUE, show_eval = TRUE, show_viz = TRUE,
-                        eval_order = NULL, viz_order = NULL, use_icons = TRUE,
+                        eval_order = NULL, viz_order = NULL,
+                        eval_cache = ".rds", viz_cache = ".rds",
+                        use_icons = TRUE,
                         quiet = TRUE, verbose = 2, ...) {
 
   if (missing(experiment) && missing(save_dir)) {
@@ -269,7 +282,9 @@ render_docs <- function(experiment, save_dir, write_rmd = FALSE,
     sim_name = title, sim_path = save_dir, author = author,
     write_filename = sprintf("%s.Rmd", output_file),
     show_code = show_code, show_eval = show_eval, show_viz = show_viz,
-    eval_order = eval_order, viz_order = viz_order, use_icons = use_icons,
+    eval_order = eval_order, viz_order = viz_order,
+    eval_cache = eval_cache, viz_cache = viz_cache,
+    use_icons = use_icons,
     use_vmodern = use_vmodern, write = write_rmd, verbose = verbose
   )
 
diff --git a/R/experiment.R b/R/experiment.R
index c4fe5b63..b0734cd9 100644
--- a/R/experiment.R
+++ b/R/experiment.R
@@ -2005,7 +2005,7 @@ Experiment <- R6::R6Class(
     #' @param ... Additional arguments to pass to [ggplot2::ggsave()].
     #'
     #' @return The `Experiment` object, invisibly.
-    export_visualizers = function(device = "pdf", width = "auto", height = "auto",
+    export_visualizers = function(device = "png", width = "auto", height = "auto",
                                   ...) {
       rlang::check_installed("ggplot2",
         reason = "to export visualizers to image.")
@@ -2020,7 +2020,6 @@ Experiment <- R6::R6Class(
       } else {
         save_dir <- private$.get_vary_across_dir()
       }
-      save_dir <- file.path(save_dir, "viz_results")
       if (!dir.exists(save_dir)) {
         dir.create(save_dir, recursive = TRUE)
       }
diff --git a/R/utils.R b/R/utils.R
index 84ffeac4..ca4c95f8 100644
--- a/R/utils.R
+++ b/R/utils.R
@@ -513,7 +513,7 @@ check_results_names <- function(names, method_name) {
 #'   `after`, and `outside` (equivalent to `before` and `end`).
 #'
 #' @returns The input `text`, but marked as HTML.
-#' @keywords internal
+#' @export
 HTML <- function(text, ..., .noWS = NULL) {
   htmlText <- c(text, as.character(rlang::dots_list(...)))
 
diff --git a/_pkgdown.yml b/_pkgdown.yml
index 9155cf8c..3b95587a 100644
--- a/_pkgdown.yml
+++ b/_pkgdown.yml
@@ -40,6 +40,7 @@ reference:
       - init_docs
       - render_docs
       - create_rmd
+      - HTML
   - title: "DGP"
     desc: >
       Functions for creating a DGP (data-generating process).
diff --git a/inst/rmd/results.Rmd b/inst/rmd/results.Rmd
index a8b01ba8..1fb6febb 100644
--- a/inst/rmd/results.Rmd
+++ b/inst/rmd/results.Rmd
@@ -31,6 +31,12 @@ params:
   viz_order:
     label: "Order of Visualizers:"
     value: NULL
+  eval_cache:
+    label: "Evaluators Cache Type:"
+    value: ".rds"
+  viz_cache:
+    label: "Visualizers Cache Type:"
+    value: ".rds"
   use_icons:
     label: "Use Icons:"
     value: TRUE
@@ -338,18 +344,119 @@ show_recipe <- function(field_name = c(
 }
 
 #' Reads in file if it exists and returns NULL if the file does not exist
+#' 
+#' @description If filetype is ".rds", the file is read in using readRDS(). 
+#'   Otherwise, the file is read in using data.table::fread().
 #'
-#' @param filename name of .rds file to try reading in
-#' @return output of filename.rds if the file exists and NULL otherwise
-get_results <- function(filename) {
+#' @param filename name of file (with file extension) to try reading in
+#' @return output of filename if the file exists and NULL otherwise
+get_results <- function(filename, filetype = ".rds") {
   if (file.exists(filename)) {
-    results <- readRDS(filename)
+    if (filetype == ".rds") {
+      results <- readRDS(filename)
+    } else {
+      results <- data.table::fread(results)
+    }
   } else {
     results <- NULL
   }
   return(results)
 }
 
+#' Reads in list of image result files
+#' 
+#' @param filenames vector of filenames (without file extension) to read in
+#' @param filetype file extension
+#' @return list of image results, wrapped as knitr::include_graphics() objects
+get_image_results <- function(filenames, filetype) {
+  purrr::map(
+    filenames,
+    function(f) {
+      if (file.exists(sprintf("%s%s", f, filetype))) {
+        return(knitr::include_graphics(sprintf("%s%s", f, filetype)))
+      } else {
+        return(NULL)
+      }
+    }
+  ) |> 
+    setNames(basename(filenames)) |> 
+    purrr::compact()
+}
+
+#' Get results from experiment
+#' 
+#' @param dir_name name of directory
+#' @param show_eval logical; whether or not to show evaluators
+#' @param show_viz logical; whether or not to show visualizers
+#' @param eval_cache file extension for cached evaluator results to read in.
+#'   Typically ".rds" or "none", but can be any file extension where
+#'   evaluator results are stored as `eval_results.ext` and can be read in
+#'   using `data.table::fread(eval_results.ext)`. If "none", evaluator results
+#'   are computed using the experiment via `evaluate_experiment()`.
+#' @param viz_cache file extension for cached visualizer results to read in.
+#'   Typically ".rds" or "none", but can be any (image) file extension 
+#'   (e.g., "png", "jpg") where the visualizer results have been previously 
+#'   stored as separate `{visualizer_name}.ext` images (e.g., using
+#'   `export_visualizers()`). If "none", visualizer results are computed using
+#'   the experiment via `visualize_experiment()`.
+get_exp_results <- function(dir_name, 
+                            show_eval = TRUE, show_viz = TRUE,
+                            eval_cache = ".rds", viz_cache = ".rds") {
+  
+  exp_fname <- file.path(dir_name, "experiment.rds")
+  fit_fname <- file.path(dir_name, "fit_results.rds")
+  eval_fname <- file.path(dir_name, sprintf("eval_results%s", eval_cache))
+  viz_fname <- file.path(dir_name, sprintf("viz_results%s", viz_cache))
+
+  exp <- get_results(exp_fname)
+  fit_results <- NULL
+  eval_results <- NULL
+  viz_results <- NULL
+  if ((eval_cache != "none") && (viz_cache == ".rds")) {
+    if (show_eval) {
+      eval_results <- get_results(eval_fname, eval_cache)
+    }
+    if (show_viz) {
+      viz_results <- get_results(viz_fname, viz_cache)
+    }
+  } else {
+    if (show_eval) {
+      if (eval_cache == "none") {
+        fit_results <- get_results(fit_fname)
+        eval_results <- evaluate_experiment(exp, fit_results)
+      } else {
+        eval_results <- get_results(eval_fname, eval_cache)
+      }
+    }
+    if (show_viz) {
+      if (viz_cache == ".rds") {
+        viz_results <- get_results(viz_fname, viz_cache)
+      } else if (viz_cache == "none") {
+        if (is.null(fit_results)) {
+          fit_results <- get_results(fit_fname)
+        }
+        if (is.null(eval_results)) {
+          eval_results <- get_results(eval_fname)
+          if (is.null(eval_results)) {
+            eval_results <- evaluate_experiment(exp, fit_results)
+          }
+        }
+        viz_results <- visualize_experiment(exp, fit_results, eval_results)
+      } else {
+        viz_results <- get_image_results(
+          file.path(dir_name, names(exp$get_visualizers())), viz_cache
+        )
+      }
+    }
+  }
+  results <- list(
+    exp = exp,
+    eval_results = eval_results,
+    viz_results = viz_results
+  )
+  return(results)
+}
+
 #' Displays output (both from evaluate() and visualize()) from saved results under
 #' a specified directory
 #'
@@ -428,14 +535,26 @@ show_results <- function(dir_name, depth, base = FALSE, show_header = TRUE,
     code_label <- "Varying Parameters"
   }
 
-  exp_fname <- file.path(dir_name, "experiment.rds")
-  eval_fname <- file.path(dir_name, "eval_results.rds")
-  viz_fname <- file.path(dir_name, "viz_results.rds")
-
-  exp <- get_results(exp_fname)
-  eval_results <- get_results(eval_fname)
-  viz_results <- get_results(viz_fname)
+  exp_results <- get_exp_results(
+    dir_name, 
+    show_eval = params$show_eval, show_viz = params$show_viz,
+    eval_cache = params$eval_cache, viz_cache = params$viz_cache
+  )
+  exp <- exp_results$exp
+  eval_results <- exp_results$eval_results
+  viz_results <- exp_results$viz_results
 
+  if (write_flag && (params$show_eval || params$show_viz)) {
+    results <- sprintf(
+      "results <- get_exp_results(\n  '%s', show_eval = %s, show_viz = %s, eval_cache = '%s', viz_cache = '%s'\n)",
+      dir_name, 
+      params$show_eval, params$show_viz,
+      params$eval_cache, params$viz_cache
+    ) |>
+      write_code_chunk(chunk_args = "results = 'asis'") |>
+      write(old_text = results, write_flag = write_flag)
+  }
+  
   if (!is.null(eval_results) && params$show_eval) {
     results <- write(
       sprintf(showtype_template, evaluator_label),
@@ -471,7 +590,7 @@ show_results <- function(dir_name, depth, base = FALSE, show_header = TRUE,
         }
         if (write_flag) {
           results <- sprintf(
-            "show_results('%s', '%s', 'evaluator')", dir_name, eval_name
+            "show_results(results, '%s', 'evaluator')", eval_name
           ) |>
             write_code_chunk(chunk_args = "results = 'asis'") |>
             write(old_text = results, write_flag = write_flag)
@@ -522,7 +641,8 @@ show_results <- function(dir_name, depth, base = FALSE, show_header = TRUE,
           plt <- plts[[plt_name]]
           is_plot <- inherits(plt, "plotly") || 
             inherits(plt, "gg") || 
-            inherits(plt, "ggplot")
+            inherits(plt, "ggplot") ||
+            inherits(plt, "knit_image_paths")
           
           if (params$use_vmodern && is_plot) {
             chunk_args <- "fig.height = %s, fig.width = %s, out.width = '100%%', add.panel = TRUE"
@@ -534,7 +654,7 @@ show_results <- function(dir_name, depth, base = FALSE, show_header = TRUE,
           
           if (write_flag) {
             results <- sprintf(
-              "show_results('%s', '%s', 'visualizer')", dir_name, viz_name
+              "show_results(results, '%s', 'visualizer')", viz_name
             ) |>
               write_code_chunk(
                 chunk_args = sprintf(
@@ -567,9 +687,7 @@ show_results <- function(dir_name, depth, base = FALSE, show_header = TRUE,
         old_text = results, write_flag = write_flag
       )
       if (write_flag) {
-        results <- sprintf(
-          "show_results('%s', NULL, 'vary_params')", dir_name
-        ) |>
+        results <- c("show_results(results, NULL, 'vary_params')") |>
           write_code_chunk(chunk_args = "max.height='200px'") |>
           write(old_text = results, write_flag = write_flag)
       } else {
diff --git a/inst/rmd/results_header_template.Rmd b/inst/rmd/results_header_template.Rmd
index abbf97e5..42b8e096 100644
--- a/inst/rmd/results_header_template.Rmd
+++ b/inst/rmd/results_header_template.Rmd
@@ -21,6 +21,8 @@ options(
   knitr.kable.NA = 'NA'
 )
 
+library(simChef)
+
 local({
   hook_output <- knitr::knit_hooks$get("output")
   knitr::knit_hooks$set(
@@ -69,19 +71,120 @@ get_descendants <- function(dir_name) {
   return(experiments)
 }
 
-#' Reads in file if it exists and returns NULL if the file does not exist.
+#' Reads in file if it exists and returns NULL if the file does not exist
+#' 
+#' @description If filetype is ".rds", the file is read in using readRDS(). 
+#'   Otherwise, the file is read in using data.table::fread().
 #'
-#' @param filename Name of .rds file to try reading in.
-#' @return Output of filename.rds if the file exists and NULL otherwise.
-get_results <- function(filename) {
+#' @param filename name of file (with file extension) to try reading in
+#' @return output of filename if the file exists and NULL otherwise
+get_results <- function(filename, filetype = ".rds") {
   if (file.exists(filename)) {
-    results <- readRDS(filename)
+    if (filetype == ".rds") {
+      results <- readRDS(filename)
+    } else {
+      results <- data.table::fread(results)
+    }
   } else {
     results <- NULL
   }
   return(results)
 }
 
+#' Reads in list of image result files
+#' 
+#' @param filenames vector of filenames (without file extension) to read in
+#' @param filetype file extension
+#' @return list of image results, wrapped as knitr::include_graphics() objects
+get_image_results <- function(filenames, filetype) {
+  purrr::map(
+    filenames,
+    function(f) {
+      if (file.exists(sprintf("%s%s", f, filetype))) {
+        return(knitr::include_graphics(sprintf("%s%s", f, filetype)))
+      } else {
+        return(NULL)
+      }
+    }
+  ) |> 
+    setNames(basename(filenames)) |> 
+    purrr::compact()
+}
+
+#' Get results from experiment
+#' 
+#' @param dir_name name of directory
+#' @param show_eval logical; whether or not to show evaluators
+#' @param show_viz logical; whether or not to show visualizers
+#' @param eval_cache file extension for cached evaluator results to read in.
+#'   Typically ".rds" or "none", but can be any file extension where
+#'   evaluator results are stored as `eval_results.ext` and can be read in
+#'   using `data.table::fread(eval_results.ext)`. If "none", evaluator results
+#'   are computed using the experiment via `evaluate_experiment()`.
+#' @param viz_cache file extension for cached visualizer results to read in.
+#'   Typically ".rds" or "none", but can be any (image) file extension 
+#'   (e.g., "png", "jpg") where the visualizer results have been previously 
+#'   stored as separate `{visualizer_name}.ext` images (e.g., using
+#'   `export_visualizers()`). If "none", visualizer results are computed using
+#'   the experiment via `visualize_experiment()`.
+get_exp_results <- function(dir_name, 
+                            show_eval = TRUE, show_viz = TRUE,
+                            eval_cache = ".rds", viz_cache = ".rds") {
+  
+  exp_fname <- file.path(dir_name, "experiment.rds")
+  fit_fname <- file.path(dir_name, "fit_results.rds")
+  eval_fname <- file.path(dir_name, sprintf("eval_results%s", eval_cache))
+  viz_fname <- file.path(dir_name, sprintf("viz_results%s", viz_cache))
+
+  exp <- get_results(exp_fname)
+  fit_results <- NULL
+  eval_results <- NULL
+  viz_results <- NULL
+  if ((eval_cache != "none") && (viz_cache == ".rds")) {
+    if (show_eval) {
+      eval_results <- get_results(eval_fname, eval_cache)
+    }
+    if (show_viz) {
+      viz_results <- get_results(viz_fname, viz_cache)
+    }
+  } else {
+    if (show_eval) {
+      if (eval_cache == "none") {
+        fit_results <- get_results(fit_fname)
+        eval_results <- evaluate_experiment(exp, fit_results)
+      } else {
+        eval_results <- get_results(eval_fname, eval_cache)
+      }
+    }
+    if (show_viz) {
+      if (viz_cache == ".rds") {
+        viz_results <- get_results(viz_fname, viz_cache)
+      } else if (viz_cache == "none") {
+        if (is.null(fit_results)) {
+          fit_results <- get_results(fit_fname)
+        }
+        if (is.null(eval_results)) {
+          eval_results <- get_results(eval_fname)
+          if (is.null(eval_results)) {
+            eval_results <- evaluate_experiment(exp, fit_results)
+          }
+        }
+        viz_results <- visualize_experiment(exp, fit_results, eval_results)
+      } else {
+        viz_results <- get_image_results(
+          file.path(dir_name, names(exp$get_visualizers())), viz_cache
+        )
+      }
+    }
+  }
+  results <- list(
+    exp = exp,
+    eval_results = eval_results,
+    viz_results = viz_results
+  )
+  return(results)
+}
+
 #' Show recipe content.
 #' 
 #' @param objs List of DGP/Method/Evaluator/Visualizer objects. Should be one of
@@ -114,16 +217,12 @@ show_recipe <- function(objs, name, experiment_name,
 #' @param field_name One of "evaluator", "visualizer", or "vary_params", 
 #'   indicating whether to return the results of an Evaluator, Visualizer, or
 #'   the varying across parameters of the Experiment, respectively.
-show_results <- function(dir_name, name,
+show_results <- function(results, name,
                          field_name = c("evaluator", "visualizer", "vary_params")) {
   field_name <- match.arg(field_name)
-  exp_fname <- file.path(dir_name, "experiment.rds")
-  eval_fname <- file.path(dir_name, "eval_results.rds")
-  viz_fname <- file.path(dir_name, "viz_results.rds")
-
-  exp <- get_results(exp_fname)
-  eval_results <- get_results(eval_fname)
-  viz_results <- get_results(viz_fname)
+  exp <- results$exp
+  eval_results <- results$eval_results
+  viz_results <- results$viz_results
   
   if (field_name == "evaluator") {
     doc_nrows <- exp$get_evaluators()[[name]]$doc_nrows
diff --git a/man/Experiment.Rd b/man/Experiment.Rd
index c6437d54..f396b4d8 100644
--- a/man/Experiment.Rd
+++ b/man/Experiment.Rd
@@ -1030,7 +1030,7 @@ Export all cached \code{Visualizer} results from an
 \code{Experiment}'s results directory (see \code{\link[=get_save_dir]{get_save_dir()}}).
 \subsection{Usage}{
 \if{html}{\out{<div class="r">}}\preformatted{Experiment$export_visualizers(
-  device = "pdf",
+  device = "png",
   width = "auto",
   height = "auto",
   ...
diff --git a/man/HTML.Rd b/man/HTML.Rd
index e2c9cf1d..ffdf3f11 100644
--- a/man/HTML.Rd
+++ b/man/HTML.Rd
@@ -23,4 +23,3 @@ The input \code{text}, but marked as HTML.
 Marks the given text as HTML, which means the tag functions will
 know not to perform HTML escaping on it. Copy of \code{htmltools::HTML()}.
 }
-\keyword{internal}
diff --git a/man/render_docs.Rd b/man/render_docs.Rd
index 67dcf723..16081449 100644
--- a/man/render_docs.Rd
+++ b/man/render_docs.Rd
@@ -18,6 +18,8 @@ render_docs(
   show_viz = TRUE,
   eval_order = NULL,
   viz_order = NULL,
+  eval_cache = ".rds",
+  viz_cache = ".rds",
   use_icons = TRUE,
   quiet = TRUE,
   verbose = 2,
@@ -75,6 +77,19 @@ in the order that they were computed.}
 display. By default, the report will display the \code{Visualizer} results
 in the order that they were computed.}
 
+\item{eval_cache}{File extension of the cached evaluator results to read in.
+Typically \code{".rds"} or \code{"none"}, but can be any file extension where
+evaluator results are stored as \code{eval_results.ext} and can be read in
+using \code{data.table::fread(eval_results.ext)}. If \code{"none"}, evaluator results
+are computed using the experiment via \code{\link[=evaluate_experiment]{evaluate_experiment()}}.}
+
+\item{viz_cache}{File extension of the cached visualizer results to read in.
+Typically \code{".rds"} or \code{"none"}, but can be any (image) file extension
+(e.g., \code{"png"}, \code{"jpg"}) where the visualizer results have been previously
+stored as separate \verb{\{visualizer_name\}.ext} images (typically using
+\code{\link[=export_visualizers]{export_visualizers()}}). If \code{"none"}, visualizer results are computed using
+the experiment via \code{\link[=visualize_experiment]{visualize_experiment()}}.}
+
 \item{use_icons}{Logical indicating whether or not to use fontawesome icons.}
 
 \item{quiet}{An option to suppress printing during rendering from knitr,
diff --git a/tests/testthat/test-docs.R b/tests/testthat/test-docs.R
index e6b5fb82..64a7532b 100644
--- a/tests/testthat/test-docs.R
+++ b/tests/testthat/test-docs.R
@@ -29,6 +29,7 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       add_evaluator(evaluator, "Evaluator") |>
       add_visualizer(visualizer, "Plot")
     results <- base_experiment$run(save = TRUE, verbose = 0)
+    export_visualizers(base_experiment)
 
     child1 <- create_experiment(
       name = "child1",
@@ -36,12 +37,14 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       save_dir = file.path(base_experiment$get_save_dir(), "child1")
     )
     results <- child1$run(save = TRUE, verbose = 0)
+    export_visualizers(child1)
     child2 <- create_experiment(
       name = "child2",
       clone_from = base_experiment,
       save_dir = file.path(base_experiment$get_save_dir(), "child2")
     )
     results <- child2$run(save = TRUE, verbose = 0)
+    export_visualizers(child2)
 
     grandchild1a <- create_experiment(
       name = "grandchild1a",
@@ -49,6 +52,7 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       save_dir = file.path(child1$get_save_dir(), "grandchild1a")
     )
     results <- grandchild1a$run(save = TRUE, verbose = 0)
+    export_visualizers(grandchild1a)
 
     grandchild1b <- create_experiment(
       name = "grandchild1b",
@@ -56,6 +60,7 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       save_dir = file.path(child1$get_save_dir(), "grandchild1b")
     )
     results <- grandchild1b$run(save = TRUE, verbose = 0)
+    export_visualizers(grandchild1b)
 
     grandchild2 <- create_experiment(
       name = "grandchild2",
@@ -63,6 +68,7 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       save_dir = file.path(child2$get_save_dir(), "grandchild2")
     )
     results <- grandchild2$run(save = TRUE, verbose = 0)
+    export_visualizers(grandchild2)
 
     greatgrandchild2 <- create_experiment(
       name = "greatgrandchild2",
@@ -70,6 +76,7 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       save_dir = file.path(grandchild2$get_save_dir(), "greatgrandchild2")
     )
     results <- greatgrandchild2$run(save = TRUE, verbose = 0)
+    export_visualizers(greatgrandchild2)
 
     expect_error(render_docs(base_experiment, verbose = 0), NA)
     expect_error(
@@ -86,6 +93,36 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       NA
     )
 
+    # check different cache modes for evaluators and visualizers
+    expect_error(
+      render_docs(
+        base_experiment, verbose = 0,
+        eval_cache = ".rds", viz_cache = "none"
+      ),
+      NA
+    )
+    expect_error(
+      render_docs(
+        base_experiment, verbose = 0,
+        eval_cache = "none", viz_cache = ".rds"
+      ),
+      NA
+    )
+    expect_error(
+      render_docs(
+        base_experiment, verbose = 0,
+        eval_cache = "none", viz_cache = "none"
+      ),
+      NA
+    )
+    expect_error(
+      render_docs(
+        base_experiment, verbose = 0,
+        eval_cache = "none", viz_cache = ".png"
+      ),
+      NA
+    )
+
     # check ... arguments to rmarkdown::render()
     expect_error(
       render_docs(
@@ -162,6 +199,30 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
     #   ),
     #   NA
     # )
+    expect_error(
+      render_docs(
+        base_experiment, write_rmd = TRUE, verbose = 0,
+        output_file = file.path(base_experiment$get_save_dir(), "test3"),
+        eval_cache = "none", viz_cache = "none"
+      ),
+      NA
+    )
+    expect_error(
+      rmarkdown::render(file.path(base_experiment$get_save_dir(), "test3.Rmd")),
+      NA
+    )
+    expect_error(
+      render_docs(
+        base_experiment, write_rmd = TRUE, verbose = 0,
+        output_file = file.path(base_experiment$get_save_dir(), "test4"),
+        eval_cache = "none", viz_cache = ".png"
+      ),
+      NA
+    )
+    expect_error(
+      rmarkdown::render(file.path(base_experiment$get_save_dir(), "test4.Rmd")),
+      NA
+    )
 
     # check defuse requirement
     output_format <- quote(rmarkdown::html_document())
diff --git a/tests/testthat/test-experiment.R b/tests/testthat/test-experiment.R
index b1e36d2f..6b0c29ce 100644
--- a/tests/testthat/test-experiment.R
+++ b/tests/testthat/test-experiment.R
@@ -983,18 +983,20 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
       add_method(method, "Method")
     fit_results <- experiment$fit(verbose = 0)
     eval_results <- experiment$evaluate(fit_results, verbose = 0)
-    fpath <- file.path(experiment$get_save_dir(), "viz_results")
+    fpath <- file.path(experiment$get_save_dir())
 
     # with no visualizers
     expect_error(experiment$export_visualizers(), NA)
-    expect_equal(length(list.files(path = fpath)), 0)
+    expect_equal(
+      length(list.files(path = fpath, pattern = ".png")), 0
+    )
 
     # warning if can't export visualizer
     experiment |> add_visualizer(plotly_plot, name = "plotly")
     viz_results <- experiment$visualize(fit_results, eval_results,
                                         save = TRUE, verbose = 0)
     expect_warning(experiment |> export_visualizers())
-    expect_equal(length(list.files(path = fpath)), 0)
+    expect_equal(length(list.files(path = fpath, pattern = ".png")), 0)
 
     # with one working visualizer
     experiment |>
@@ -1003,7 +1005,7 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
     viz_results <- experiment$visualize(fit_results, eval_results,
                                         save = TRUE, verbose = 0)
     expect_warning(expect_warning(experiment$export_visualizers()))
-    expect_equal(list.files(path = fpath), "ggplot.pdf")
+    expect_equal(list.files(path = fpath, pattern = ".png"), "ggplot.png")
 
     # using vary across
     experiment |>
@@ -1012,9 +1014,10 @@ withr::with_tempdir(pattern = "simChef-test-checkpointing-temp", code = {
     expect_warning(expect_warning(experiment$export_visualizers()))
     expect_equal(
       list.files(
-        path = file.path(experiment$get_save_dir(), "DGP", "Varying y", "viz_results")
+        path = file.path(experiment$get_save_dir(), "DGP", "Varying y"),
+        pattern = ".png"
       ),
-      "ggplot.pdf"
+      "ggplot.png"
     )
   })