Home / Batch PDF
Batch PDF
Apply one operation to many PDFs or whole folders and download the results in a ZIP, with a report that says what happened to each file, without any of them leaving your computer.
Open in the app The app opens with this tool already selected. Your PDF is not uploaded anywhere.
How it works
Add files or drag in a whole folder (its subfolders are kept inside the ZIP), choose an operation (compress, rotate, number pages, text watermark, remove metadata, protect with a password, remove the password, OCR, flatten forms or repair) and set its options. The same options apply to every file.
Files are processed one at a time. A file that fails, is damaged, is protected with a different password or is signed does not stop the batch: it appears in the report with its reason and the rest carry on. Signed files are skipped unless you tick the box that allows them and accept that their signatures will stop being valid. You can cancel at any moment and you get the ZIP of what is already finished.
When it finishes you have a ZIP (or separate files, if you prefer) and a report in the panel, as a CSV and as batch-report.txt inside the ZIP. Each file is listed as processed, unchanged, skipped, failed or canceled; the summary never says everything went well if any file was not processed. Your originals are only read, never changed.
What is not preserved, and limits
- The real limit is the memory of your device and the total size of the files, not how many files there are: each file is opened and rewritten one at a time (it needs about 3 to 4 times its size while it is processed) and every finished result is kept until you download the ZIP. A very large batch can slow down or stop the page; run it in smaller groups or choose separate files so nothing is kept.
- The ZIP does not use the ZIP64 format: when the results pass 1 GiB they are split into numbered parts and you must download every part. A single result of more than about 4 GiB does not fit and is marked as a failure.
- A file over 600 MB is refused on its own (with its row in the report) and the rest of the batch goes on.
- Digitally signed files are skipped by default. If you tick the box to process them, the report says what happened to each signature: after operations that only append to the end of the file, a viewer will usually warn that the document was modified after signing; after operations that rewrite the file, the signature no longer validates. OCR never processes signed files.
- Password-protected files need the password you type: it is one password for the whole batch. A file that password does not open is marked as a failure and the batch goes on. The password lives only in the memory of the page and is written neither into the report nor into the ZIP.
- OCR is slow (one page after another, one file after another) and a cancel waits for the page in progress to finish. It refuses signed files and password-protected files. Its quality is that of automatic text recognition: it can make mistakes.
- Removing metadata rewrites the file without its title, author, dates, XMP data, thumbnails and other hidden information, but it does NOT touch the visible text of the document, the values of form fields, the authors of annotations, embedded files, the dates of the file on your disk or the EXIF data an image may carry inside. A digital signature stops being valid.
- Compress can leave a file unchanged: only images are recompressed, and if the result does not save at least 1% the original is kept and reported as “unchanged”.
- A damaged file is not repaired unless the operation you chose is Repair: with any other operation it is marked as a failure and you are asked to repair it first. A result that was only partly repaired is flagged as such in the report and may not be the complete document.
- The options are the same for every file: rotate uses the same pages in each file (an open range such as “7-” is not accepted), the numbering starts again in every file and the watermark is text only, with no image.
- Each result is checked before it enters the ZIP with our own readers (it starts with %PDF, opens without any repair and keeps its page count), written by the same team: that does not prove it opens in every program. Tested with pikepdf, qpdf and poppler; it has not been tried in Adobe Acrobat or Reader, and only in Chromium (dragging in folders depends on the browser).
- Nothing is uploaded and nothing is stored: the files, the passwords and the results live only in the memory of the page and are lost when you close or reload it, so download the ZIP before you leave.
Why it is different
This tool works inside your browser. The file you choose stays on your device and is not sent to any server. That is why you do not need an account or an email to use it.
Other tools
- Merge PDF
- Split PDF
- Rotate PDF
- JPG to PDF
- PDF to JPG
- Compress PDF
- Sign PDF
- OCR PDF
- Unlock PDF
- Protect PDF
- Add page numbers to a PDF
- Watermark PDF
- Crop PDF
- Organize PDF pages
- Extract pages from a PDF
- Extract images from a PDF
- Extract text from a PDF
- Flatten forms and remove annotations
- Redact PDF
- Annotate PDF
- Fill PDF form
- Compare PDF
- Repair PDF