diff --git a/CRAN-SUBMISSION b/CRAN-SUBMISSION
deleted file mode 100644
index 8c4b362..0000000
--- a/CRAN-SUBMISSION
+++ /dev/null
@@ -1,3 +0,0 @@
-Version: 0.4.7
-Date: 2025-01-07 21:48:22 UTC
-SHA: 099855690ae05afaf95090f4bc9645fa2c0b9246
diff --git a/NEWS.md b/NEWS.md
index e79a855..1fa4e50 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -20,6 +20,8 @@
 
 -   Wrapped a bunch of examples from being run on CRAN submission to avoid Debian system NOTE.
 
+-   Reduced computational time for vignettes to to avoid Debian system NOTE on CRAN submission.
+
 # joineRML 0.4.6
 
 ## Housekeeping
diff --git a/cran-comments.md b/cran-comments.md
index 2732086..9e44c78 100644
--- a/cran-comments.md
+++ b/cran-comments.md
@@ -1,7 +1,6 @@
 ## Other notes
 
 * Addresses all NOTES from CRAN Package Check Results.
-* I have stopped several examples from running in order to avoid the NOTE on the Debian system. I cannot reproduce these, but am happy to wrap in `\dontrun{}` blocks.
 
 ## Test environments
 
@@ -18,8 +17,6 @@
 
 Win-Builder NOTE: "checking CRAN incoming feasibility ... NOTE"
 
-Debian NOTE: "Re-building vignettes had CPU time 5.5 times elapsed time" -- I cannot reproduce this. It passes every other platform. I suspect the qpdf compression tool is the cause, but that was requested by CMD tests.
-
 ## Reverse dependencies
 
 We checked 3 reverse dependencies, comparing R CMD check results across CRAN and dev versions of this package.
diff --git a/vignettes/joineRML-tidy.Rmd b/vignettes/joineRML-tidy.Rmd
index 0701357..b9ee3f0 100644
--- a/vignettes/joineRML-tidy.Rmd
+++ b/vignettes/joineRML-tidy.Rmd
@@ -55,7 +55,7 @@ Further to that, we only select the first 50 individuals to speed up these examp
 hvd <- hvd[hvd$num <= 50, ]
 ```
 
-```{r hvd_model_fit}
+```{r hvd_model_fit, cache = TRUE}
 set.seed(12345)
 fit <- mjoint(
   formLongFixed = list(
diff --git a/vignettes/joineRML.Rmd b/vignettes/joineRML.Rmd
index 791941b..34bd52c 100644
--- a/vignettes/joineRML.Rmd
+++ b/vignettes/joineRML.Rmd
@@ -64,6 +64,12 @@ hvd <- heart.valve[!is.na(heart.valve$grad) & !is.na(heart.valve$lvmi), ]
 
 Strictly speaking, this is not necessary because `joineRML` can handle the situation of different measurement schedules *within* subjects That is, a subject does not need to have all multiple longitudinal outcomes recorded at each visit. It is conceivable that some biomarkers will be measured more or less frequently than others. For example, invasive measurements may only be recorded annually, whereas a simple biomarker measurement might be recorded more frequently. `joineRML` can handle this situation by specifying each longitudinal outcome its own data frame.
 
+Further to that, we only select the first 50 individuals to speed up these examples:
+
+```{r hvd_data_small}
+hvd <- hvd[hvd$num <= 50, ]
+```
+
 ## Model fitting
 
 The main function in the `joineRML` package is the `mjoint()` function. Its main (required) arguments are: