I have a range of PDF files 1.pdf, 2.pdf, etc. that I would like to merge into one file, with all the PDFs tiled on one page.

Currently, I have tried pdftk to merge these files, but they are put on separate pages:

pdftk 1.pdf 2.pdf ... cat output merged.pdf

Is there a way to, instead, tile the individual PDF files into one master page on merged.pdf?


10 Answers 10


I do not know if pdftk can do this, but pdfjam can. It can be installed on debian or derivates with sudo apt install texlive-extra-utils. Then you can do:

pdfjam Page1.pdf Page2.pdf --nup 2x1 --landscape --outfile Page1+2.pdf

It puts 2 pages on one page side by side. With --nup 1x2 --no-landscapeyou can put two pages on one page on top of each other.

  • If you wanted no auto-scaling of pages (mine stretches output to fill a4 by default), use --noautoscale true Commented Mar 12, 2018 at 14:14

Not sure what you mean by tiled on one page. I was looking for a way to merge multiple PDFs on one page - on top of another. This can be done with pdftk like this:

pdftk foreground.pdf background background.pdf output merged.pdf

  • 6
    Absolutely not an answer to the question. The adjective tiled is defined as "arrange without overlap". Commented Aug 25, 2021 at 2:09
  • @ChristianHerenz it's not the reason why the answer is not good. It could be a step to the solution. But turns out that it does nothing other than concatenating the pages. Commented Feb 8, 2022 at 10:33

The pdfLaTeX based pdfnup might work for you. If you have lots of pdf-files you may need to make a long pipe of pdfjam or run it several times.

There is also pdfnup in python.


you can use montage from ImageMagick

$ montage *.pdf merged.pdf

see also http://www.imagemagick.org/script/montage.php

  • 18
    ImageMagick montage will not handle vectorized images and fonts properly. As a result, the output from the montage command might look blurry. See also the explanation here: superuser.com/a/479767/149568 Commented Feb 20, 2013 at 13:00
  • 3
    This does NOT answer the question, which is asking about pdftk
    – Zimba
    Commented Sep 28, 2019 at 12:14

This script will tile the pdf pages for you. Change the slice to what you need per page.


latexhead = <<'EOF'
latextail = <<'EOF'

pages = %x[pdfinfo #{ARGV[0]}].split(/\n/).select{|x| x=~ /Pages:/}[0].split(/\s+/)[1].to_i
puts latexhead
s = (1..pages).each_slice(4).to_a
s.each do |a|
  puts "\\begin{figure}"
  a.each do |p|
    puts "\\includegraphics[page=#{p},scale=0.4,width=.5\\textwidth]{#{ARGV[0]}}"
  puts "\\end{figure}"
puts latextail

If filenames are in "system specific" order, then pdftk *.pdf cat output merged.pdf should work just fine.

Here is what I mean by "system specific" order.

I've got 3 files on my Ubuntu 11.04: 1.pdf, 2.pdf, 10.pdf
Files are merged in order: 10.pdf 1.pdf 2.pdf (ls -l returned the same order as in merged file)

Safest naming convention: 0001.pdf, 0002.pdf, etc.

  • 4
    As I stated, this command creates a multi-page PDF. As stated in my question, I am looking for the method of using pdftk to make a one-page PDF that contains input PDFs as tiles. Commented Dec 11, 2011 at 9:35
  • Combine my answer with @micke's. I haven't checked whether pdfnup can take name with wildcard (*.pdf) as argument, but you can use pdf generated by pdftk. Check pdfnup or pdfjam with --nup option.
    – szemek
    Commented Dec 11, 2011 at 16:49
  • Unless you quote them, wildcards like * are expanded by the shell you're using. pdfnup doesn't ever see *.pdf, instead it sees a list of every file in the working directory with filenames ending in .pdf.
    – evilsoup
    Commented Aug 27, 2013 at 16:11

If you've got a large number of PDFs in one folderstructure, and you've got a TeX-Installation, this script puts all the PDFs recursivly into one large file:

# pdfdir OUTPUT_FILE
# produces one big PDF file of all PDF files in .
if [ $# -ne 1 ] || [ -z "$1" ]; then
  echo "Syntax: pdfdir OUTPUT_FILE"
  exit 1
FILE="$(echo "$1"|sed -e 's/\.\(pdf\|tex\)$//')"
for F in "$FILE" "$FILE.tex" "$FILE.pdf" "$FILE.aux" "$FILE.log" ; do
  if [ -e "$F" ]; then
    echo "$F exists already."
    exit 2
cat >"$FILE.tex" <<EOF
# helper functions
exist_pdf_files () {
  [ $(find -L "$1" -name \*.pdf -o -name \*.PDF -type f 2>/dev/null|wc -l) -eq 0 ] && return 1
  return 0
list_directories () {
  find -L "$1" -maxdepth 1 -mindepth 1 -type d 2>/dev/null | sort
list_pdf_files () {
  # version with " around filenames:
  #find -L "$1" -maxdepth 1 -mindepth 1 -name \*.pdf -o -name \*.PDF -type f 2>/dev/null | sort | \
  #  sed -e 's/^/\\includepdf[pages=-]{"/; s/$/"}%/'
  # version without " around filenames:
  find -L "$1" -maxdepth 1 -mindepth 1 -name \*.pdf -o -name \*.PDF -type f 2>/dev/null | sort | \
    sed -e 's/^/\\includepdf[pages=-]{/; s/$/}%/'
tex_headline () {
    echo "$1" | sed -e 's/_/\\_/g'
# current folder (lefel 0):
list_pdf_files . >>"$FILE.tex"
# Bearbeite Ebene 1:
list_directories . | while read -r DIR1; do
  # Are there PDFs in folders below that level?
  exist_pdf_files "$DIR1" || continue
  # Yes ...
  tex_headline "\section{${DIR1##*/}}%"
  # those:
  list_pdf_files "$DIR1"
  # Level 2:
  list_directories "$DIR1" | while read -r DIR2; do
    exist_pdf_files "$DIR2" || continue
    tex_headline "\subsection{${DIR2##*/}}%"
    list_pdf_files "$DIR2"
    # Level 3:
    list_directories "$DIR2" | while read -r DIR3; do
      exist_pdf_files "$DIR3" || continue
      tex_headline "\subsubsection{${DIR3##*/}}%"
      list_pdf_files "$DIR3"
      # Level 4:
      list_directories "$DIR3" | while read -r DIR4; do
        exist_pdf_files "$DIR4" || continue
        tex_headline "\paragraph{${DIR4##*/}}%"
        list_pdf_files "$DIR4"
        # Level 5:
        list_directories "$DIR4" | while read -r DIR5; do
          exist_pdf_files "$DIR5" || continue
          tex_headline "\subparagraph{${DIR5##*/}}%"
          list_pdf_files "$DIR5"
done >>"$FILE.tex"
echo "\end{document}%" >>"$FILE.tex"
echo "Sourcecode to PDF directly [J/n]"
read -r ANSWER
case "$ANSWER" in
[JjYy]) ;;
*) exit 0 ;;
pdflatex "$FILE"
[ $? -eq 0 ] && rm -f "$FILE.aux" "$FILE.log" "$FILE.tex"

I have not written that code, I've got it from a discussion here: http://www.listserv.dfn.de/cgi-bin/wa?A2=ind1201&L=tex-d-l&T=0&P=10771

It is very usefull. I've translated some German comments into English.

Regards, Alexander


This link: http://www.verypdf.com/wordpress/201302/how-to-combine-4-page-pdf-into-1-page-pdf-file-to-save-ink-and-papers-34505.html shows VeryPDF PDF Stitcher can do the job vertically or sideways. Download at: http://www.verypdf.com/app/pdf-stitch/index.html

  1. Save pages wanted into a pdf doc with something like Acrobat Pro.

  2. Print document using the multiple pages to a page feature, back to a pdf document.

Multiple pages to a single page. :-)

  • 1
    I see your answer is downvoted without a comment. That's not cool, but there are certainly ways to improve your answer. Improve its grammar, add a description to the top as to why you are doing this, etc.
    – user477799
    Commented Feb 13, 2017 at 7:49
  • upvoted this answer - only because downvotes without a comment are not cool IMO! Commented Nov 14, 2017 at 0:45
  • 3
    Downvoted because the subject says "with pdftk". Your answer is maybe not off topic, but clearly not what the questioner is looking for.
    – Nik
    Commented Dec 18, 2018 at 12:22
  • I agree with @Nik's conjecture.
    – stephanmg
    Commented Jul 8, 2020 at 9:11

Sorry for not answering your specific question about pdftk, but since other answers exist show alternatives, I thought I'd chime in and present yet another alternative.

With a system that has poppler installed, you should have the pdfunite program. Using it is simple:

pdfunite foo.pdf bar.pdf   /tmp/output.pdf
  • 1
    This does not tile anything onto one page. Commented Jan 23, 2022 at 0:10
  • true. I'm not the only one having misunderstood the question, though. I guess it could be clarified by putting something pertaining to tiling in the title of the question. Commented May 30 at 12:27

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .