HorizontalCuts across the page into rows (N cuts = N+1 rows)
VerticalCuts down the page into columns (M cuts = M+1 columns)
Result
Apply to
Originalswhen unchecked, the pieces are placed directly after the original page
Output
Converting to images makes the text unselectable and unsearchable, and the file noticeably larger — what you get in return is that the content you cut away no longer appears in the file at all. Pages that were not cut are kept as they were; they have nothing hidden in them.
Preview (the red dashed lines are the cuts — drag them to fine-tune)
Page 1
Processing…
How to use: 1. A horizontal cut runs along a horizontal line, dividing the page into rows; a vertical cut runs down the page, dividing it into columns. Both can be used together. 2. The number of cuts in each direction is exactly that — a count of cuts, so N cuts give N+1 pieces. N horizontal cuts combined with M vertical ones produce (N+1)×(M+1) pieces; one cut each way, for example, gives 4. 3. Enter 0 for a direction to leave it uncut — but the two cannot both be 0. 4. The cut lines divide the page evenly to begin with; drag a red dashed line in the preview to nudge one. Every page keeps its own positions, and a single click copies the current page's positions to all of them. 5. “Custom pages” accepts individual pages and ranges, for example 1-3,5,8; page numbers start at 1. 6. By default the page that was cut is removed; untick the box and it is kept, with its pieces following straight after it. 7. “Keep the text (crop)” only changes the visible area of the page — what you cut away is still inside the file. To really get rid of it, choose “Remove completely (to images)”, at the cost of text that is no longer selectable or searchable. 8. Everything is processed locally in your browser — your PDF is never uploaded.
Upload a PDF and cut the content of each page horizontally (top / bottom) and vertically (left / right) into equal parts. Both directions can be used together — one cut each turns a page into four. Commonly used for two-up book scans, A3 impositions and two-column handouts. The cut lines can be dragged page by page in the preview, which suits scans whose spine sits off-centre. You can limit it to a page range and choose whether to keep the original page; the default cropping mode keeps the text selectable, and a “Remove completely” mode takes the cut-away content out of the file for good. All processing happens locally in your browser and nothing is uploaded.
About cutting one PDF page into several
When you need to cut a page open
Scans are the usual case. Scan an open book spread by spread and every page of the resulting PDF actually holds two pages side by side. Handouts and meeting material are often imposed onto A3, putting two pages of text on one sheet. Slides printed as handouts can carry four or six per sheet. These files look short, but hold two to six times as much content as the page count suggests. Cutting each page along its original layout restores one page per screen, so that reading, annotating and citing page numbers all line up again.
Splitting an A3 exam paper or imposed handout into two A4 pages
This is the most common kind of two-up page. A3 is 420 × 297 mm, so lay it landscape, cut once down the middle, and each half measures 210 × 297 mm — exactly one A4 sheet. The sizes match by design, so nothing has to be scaled. For A3 exam papers and A3-imposed handouts or meeting material, choose one vertical cut and every exported page comes out as a standard A4, ordered left half first, then right. If the material is imposed top-and-bottom instead, switch to one horizontal cut. When you print, pick “Actual size” or “100%” rather than “Fit to page”, or the printer shrinks it a second time and you end up with wide margins and small type. Scanned exam papers are rarely folded dead centre, so drag the red dashed line in the preview onto the real seam — every page can be dragged on its own.
Why two-up scans are painful on phones and e-readers
A two-up scan is usually landscape, while phones and e-ink readers are portrait. Fit the whole sheet on screen and each individual page gets half the width, far too small to read. Zoom in and you are dragging sideways for every line, then dragging back to start the next one. Once cut, each page fills the portrait screen exactly: you just turn pages, with no zooming and no panning. This is the single most worthwhile step before loading a scanned book onto a reader.
Cutting, cropping and splitting into files are three different things
Cutting slices a page's layout down the middle, turning one page into two or four: the page count goes up and none of the content is lost. Cropping trims the white margins or the running head off the edges: the page count stays the same, and what you trimmed is gone. Splitting into files divides one PDF into several separate files by page number: pages and layout are untouched, only the number of files changes. All three get called "splitting" in casual usage, but they solve different problems — use cutting when two pages are crammed into one, cropping when the margins are too wide, and the third one when you need to send a few pages on their own.
What usually comes after cutting
The result is a new PDF with the pages ordered left-then-right or top-then-bottom, ready to read as it is. If you then want to pull out a few pages to send on their own, use Merge PDF and upload just that one file, entering the pages you want to keep in the page range — the export is exactly the extract. The same tool handles joining several cut documents back into one.
Another common next step is dropping the PDF altogether. Some platforms only accept images, and in a chat window a picture gets opened far more readily than an attachment — for that, use PDF to Image to export each freshly cut page as a PNG or JPG; several pages are packed into one archive automatically. The other way round, if what you have is a stack of photos rather than a PDF, first use Image to PDF to build one, then come back and cut it.
FAQ
I want to split a PDF into several files, or extract just a few pages — can this tool do that?
No, those are two different things. This tool cuts the page itself: it slices a whole page down the middle so that one page becomes two or four, and the page count goes up. That is what you want for two-up book scans and A3 impositions. If instead you want to split a PDF by page number, or keep only pages 3 to 5, use Merge PDF: upload just that one file and enter 3-5 in its page range — the export is exactly the pages you picked.
How do I split an A3 exam paper into two A4 pages for printing?
Upload the PDF, set vertical cuts to 1, leave everything else at its default and export. An A3 sheet laid landscape is exactly two A4s (420 × 297 mm halved is 210 × 297 mm), so every page that comes out is a standard A4 and nothing is shrunk. Print with “Actual size / 100%” selected, not “Fit to page”, or it gets scaled down a second time. If the paper is imposed top-and-bottom, use one horizontal cut instead. When a scan's seam is off centre, drag the red dashed line in the preview onto the real seam — each page is adjusted separately, and if the whole file is off by the same amount, adjust one page and click “Apply this page's position to all pages”.
Does “A3 to A4” mean shrinking the content? How is that different from splitting into two?
“A3 to A4” is used for two opposite things, so decide which one you actually want. The first is scaling: the whole A3 layout is shrunk proportionally onto one A4. The page count stays the same and the type ends up at about 71% of its original size. That needs no tool at all — just set the scaling to “Fit to A4” in your print dialog. The second is splitting, which is what this tool does: one A3 becomes two A4s, the page count doubles and the type keeps its original size. Exam papers, two-column handouts and imposed material almost always want the second one, because the first usually prints too small to read. And if what you actually want is “split one PDF into several files by page number”, that is a third thing again — use Merge PDF, uploading just this one file and filling in the page range.
Does this tool upload my PDF to a server?
No. Reading the file, rendering the preview, cutting the pages and exporting the result all happen locally in your browser, and the page never uploads your PDF — so contracts, internal material and unpublished documents are safe here.
What is the difference between a horizontal and a vertical cut, and can I use both?
A horizontal cut runs along a horizontal line, separating the page into upper and lower rows — useful for pulling apart two-up layouts and imposition sheets. A vertical cut runs down the page into left and right columns, which is the usual way to turn a landscape spread (a scanned two-page book spread, say) into two separate pages. Use both together and the page is cut into a grid, rows first, then columns.
How does the number of cuts translate into the number of pieces?
The number of cuts is literally how many times the page is cut, so in one direction you get one more piece than cuts: 1 horizontal cut gives 2 rows, 2 cuts give 3 rows. With both directions set, the total is rows multiplied by columns — 1 horizontal and 1 vertical cut give 2 rows × 2 columns = 4 pieces. Enter 0 for a direction to leave it uncut.
The scanned pages are not centred — can I move the cut line off-centre?
Yes. The cut lines start out dividing the page evenly, but in the preview you can drag a red dashed line with the mouse or a finger and drop it on the real spine or text block. On a spread scan the spine is rarely dead centre, and the offset differs from sheet to sheet, which is why every page remembers its own cut lines: adjust whichever page you are looking at and the others stay put. If the whole file is offset the same way, set one page up and press “Apply to all pages”. “Reset this page” and “Reset all pages” return to an even split at any time.
Can I split only certain pages?
Yes. Choose “Custom pages” and type in the page numbers; individual pages and ranges can be mixed freely, for example 1-3,5,8 meaning pages 1, 2, 3, 5 and 8. Page numbers start at 1, and any page you don't list is left exactly as it was.
Can I keep the original pages after splitting?
Yes. By default the pieces replace the page they came from; untick “Remove the original page after splitting” and the original is kept, with its pieces immediately after it so you can compare them.
Does splitting reduce sharpness or damage the text?
It depends on the output mode. The default “Keep the text (crop)” only changes the visible area of the page. Nothing is converted to an image, the original vector graphics and selectable text are all preserved, and the result is exactly as sharp as the source. With “Remove completely (to images)”, every piece is re-rendered as an image at the resolution you pick, and the text becomes pixels along with it: no longer selectable, no longer searchable, and noticeably bigger. Use the first when you want fidelity and searchable text, the second when you want the cut-away content really gone.
Is the half I cut off really deleted, or can someone recover it?
It depends on the output mode you chose, and this is worth being clear about. “Keep the text (crop)” merely shrinks the page's visible area down to the piece you wanted; the content that was cut away is still sitting in the file, intact — whoever receives the file only has to change the visible area (CropBox / MediaBox) back, or run any text-extraction tool over it, to read the other half. That is perfectly fine for pulling a two-up scan apart to read it, but it is not redaction, and it is certainly not deletion. If what you are sending out is a contract or internal material whose other half must not leak, choose “Remove completely (to images)”: the tool re-renders each piece separately as an image along the cut lines you dragged, so the region you cut away is never written to the output file at all, and the source file's metadata, bookmarks and attachments are not carried over either. The cost is text that is no longer selectable or searchable, and a larger file.
Why can some PDFs not be processed?
A PDF with an open password cannot be read, and the page will say so explicitly — remove the password protection with another tool first. Files that some software (WPS, for instance) labels “encrypted” but that open without a password are handled automatically and split as normal.
Are scans and rotated pages cut correctly?
Yes. The preview draws the cut lines in the page's actual display orientation, and the tool factors the page rotation into where it cuts — so “horizontal” and “vertical” always mean what you see in the preview.