Remove PDF Metadata
Delete a PDF's document properties, its XMP metadata packet and its file identifiers, then download a clean copy — all on your own device, with nothing uploaded.
- Free
- No account
- Runs in your browser
- Nothing uploaded
Runs entirely in your browser — your input is never uploaded, logged, or stored.Privacy policy
What is Remove PDF Metadata?
Remove PDF Metadata deletes the places a PDF most often advertises who made it and how. It clears the document information dictionary — the /Info entries defined in PDF 32000-1 §14.3.3: Title, Author, Subject, Keywords, Creator, Producer, CreationDate and ModDate — together with any non-standard keys a particular writer added, such as /Company. It then deletes the XMP metadata packet, the XML description defined by ISO 16684-1 that many applications attach to a file (often repeating the same title and author), and it removes the trailer's file identifiers (the /ID pair) that tie a document to one specific file. Unlike an editor that only blanks the eight standard fields, this tool removes the entire /Info dictionary once it is empty, so the saved file has no document properties at all, and it writes no neutral Producer string back in their place.
Drop a PDF and it is read on the spot: the panel lists what was found in each of the three places before anything is changed. Click Remove metadata and the tool deletes those entries, re-saves the document and offers <name>-no-metadata.pdf to download. Your original file is not changed. If the PDF carries none of this metadata, the tool says so and offers no download, because there is nothing to remove.
What counts as "metadata" here is deliberately narrow, and it is worth knowing what that leaves out. XMP can also be attached to individual pages or to embedded objects by some producing applications; this tool looks only at the catalog's /Metadata stream, which is where the XMP packet almost always lives. Content is never touched: page text and images, page sizes, links, bookmarks and form fields are all kept exactly as they were, and the field appearance streams are not redrawn. The names inside form fields, the files attached to a PDF, and metadata baked into embedded objects are not covered.
Treat the result as a strong clean-up for sharing, not as an anonymiser. pdf-lib rewrites the whole file when saving rather than appending an incremental update, so in a straightforward document the old text is not left behind as a later revision — but strings can still survive inside embedded objects, form field names and attached files, and someone determined to inspect the raw bytes of a file may still recover traces. The rewrite also means a digital signature that covered the original file stops validating. Files that are password-protected or permission-locked are refused with a message rather than being written back broken.
Everything runs in your browser tab with pdf-lib; the PDF is never uploaded, and there is no server, no API and no account. You can confirm that in your browser's developer tools (Network tab): no request carries your document. Files up to 200 MB are accepted — the whole file is held in memory, so the limit keeps the tab from running out of it.
1. The file's leading bytes are checked for the PDF signature, then pdf-lib opens it with metadata updating switched off, so the library itself never stamps the file; an encrypted or damaged file is refused with a specific message. 2. The /Info dictionary is read directly — every key it holds, not only the eight standard ones — along with the catalog's /Metadata stream (and its decoded size) and the trailer's /ID. A value stored as a number or an array reads as blank rather than failing the whole read; a date is shown in UTC. 3. Every /Info entry is deleted. When the dictionary is left empty, the dictionary object and the trailer's /Info entry are deleted too, so the saved file has no document properties rather than empty ones. 4. The catalog's /Metadata entry is deleted, and the stream object behind it is deleted as well so it cannot be written into the new file as an orphan still carrying the XML. 5. The trailer's /ID pair is deleted. It is optional and only required alongside encryption, and encrypted files never reach this step. 6. The document is re-saved with object streams on and form-field appearances left as they are, so only the metadata changes, and it downloads as <name>-no-metadata.pdf.
| Remove metadata (this tool) | Edit metadata (edit-pdf-metadata) | |
|---|---|---|
| The /Info dictionary | Deletes every entry, then the dictionary itself | Edits the eight standard entries, or writes empty values |
| XMP metadata packet | Deletes the catalog's /Metadata stream and the object behind it | Leaves it exactly as it is |
| File identifiers (/ID) | Deletes the trailer's /ID pair | Leaves them untouched |
| Producer | Left absent — no neutral string is written | Kept, or set to whatever you type |
| Output name | <name>-no-metadata.pdf | <name>-metadata.pdf |
| When to use it | Sharing a file you do not want to advertise its title, author or software | Correcting or setting the properties a reader shows in Document Properties |
Worked examples
- A PDF with all eight /Info entries, an XMP packet and file identifiers reports "10 items found". After removing, the Before column reads the eight entries and the two "present" rows while the After column reads none for all three — and the file downloads as <name>-no-metadata.pdf.
- A PDF whose /Info holds only a Title reports "1 item found". Its Document properties row changes from Title "Q3 Report" to none; the same PDF carries no XMP and no /ID, so those two rows already read none.
- A PDF whose /Info also holds a writer's own /Company key reports that entry alongside the standard ones, and it goes too — the whole dictionary is removed, not just the eight standard keys.
- A PDF that is already clean — no /Info, no XMP packet and no /ID — reports "Nothing to remove" and offers no download, because nothing would change.
- A JPG saved with a .pdf name is refused: "scan.pdf" is actually a JPG file, not PDF. The extension does not match the contents — use a JPG tool for this file.
- A password-protected locked.pdf is refused: "locked.pdf" is encrypted (password-protected or permission-locked), so its metadata cannot be removed. Open it in a PDF reader, save or print an unprotected copy, and use that.
- A 250 MB PDF is refused: "report.pdf" is 250.00 MB, over the 200.00 MB limit for this tool. Everything runs in your browser's memory, so very large files can crash the tab.
How to use Remove PDF Metadata
- Click "Choose a PDF" or drop a PDF onto the box. The file is read on your device and the panel lists the metadata found in each of the three places.
- Check the "Metadata found" list — it names the /Info entries and their values, whether an XMP packet is present and its size, and whether file identifiers are present.
- Click "Remove metadata". The tool deletes the entries, re-saves the document and shows a Before/After list so you can see exactly what changed.
- Click "Download" to save <name>-no-metadata.pdf. Your original file is untouched.
Common errors
- "scan.pdf" is actually a JPG file, not PDF. The extension does not match the contents — use a JPG tool for this file. — the file's bytes are an image, whatever the name says.
- "notes.pdf" is empty (0 bytes). Choose the file again — it may not have finished downloading. — the file arrived empty; pick it again once the download has finished.
- "locked.pdf" is encrypted (password-protected or permission-locked), so its metadata cannot be removed. Open it in a PDF reader, save or print an unprotected copy, and use that. — the tool cannot decrypt the file, so it refuses it; re-save an unprotected copy first.
- "big.pdf" is 250.00 MB, over the 200.00 MB limit for this tool. Everything runs in your browser's memory, so very large files can crash the tab. — compress or split the PDF first, then remove its metadata.
- "broken.pdf" could not be read — the PDF is damaged or incomplete. Download it again, or open it in a PDF reader and save a fresh copy. — the file's structure is broken; get a clean copy before stripping it.
- "empty.pdf" has no pages — there is nothing to strip. — the catalog has no pages to save, so there is nothing to work with.
- The title came back after I saved in another app — that app wrote its own metadata, including possibly its own Producer string, when it re-saved the file. Remove the metadata last, or repeat the removal in that app.
- A digital signature stopped validating — saving rewrites the file, and any change to a signed PDF breaks the signature because it covered the original bytes. Keep a signed copy if you need the signature to hold.
FAQ
How do I remove metadata from a PDF?
Drop the PDF here and click Remove metadata. The tool deletes the document information dictionary entries (/Info), the catalog's XMP metadata packet and the trailer's file identifiers, then offers <name>-no-metadata.pdf to download. The file is processed in your browser and is never uploaded.
Does this really remove everything a PDF knows about me?
No, and no browser tool can promise that. It removes the three places metadata most often lives — /Info, the catalog's XMP packet and the /ID pair — and it deletes the whole /Info dictionary rather than blanking its fields. But strings can still survive inside embedded objects, form field names and attached files, XMP attached to individual pages is not searched for, and a determined inspection of the raw bytes may still recover traces. It is a strong clean-up for sharing, not an anonymiser.
What is the difference between removing metadata and editing it?
Editing changes the values: you type a new title or clear a field, and empty ones stay in the file. Removing deletes the entries themselves and then the whole /Info dictionary, plus the XMP packet and the file identifiers — so the saved file carries none of them. Use Remove PDF Metadata when you want the information gone; use the edit tool when you want to correct or set it.
Does it delete the XMP metadata packet?
Yes — the XMP packet stored in the catalog's /Metadata stream is deleted, and so is the stream object behind it, so it cannot be written back as an orphan. XMP that some producers attach to individual pages or to embedded objects is not searched for, and this tool does not look for it.
Are the file identifiers (/ID) safe to remove?
Yes. The /ID pair in the trailer is optional; it is only required alongside encryption, and encrypted files are refused here rather than modified. Removing it leaves a valid PDF that opens normally.
Will removing metadata change how my PDF looks?
No. Only metadata is deleted: page content, page sizes, images, links, bookmarks and form fields are untouched, and the form's field appearance streams are not redrawn. The file is re-saved as a whole, though, so a digital signature that covered the original bytes will no longer validate.
Is it safe to clean a confidential PDF here?
Yes. The file is read, changed and re-saved inside your browser tab with pdf-lib; it is not uploaded, and there is no server, API key or account. You can confirm in your browser's developer tools (Network tab) that no request contains your document.
Is there a file size limit?
Files up to 200 MB are accepted. Everything runs in your device's memory, and the limit keeps the tab from running out of it. Password-protected or permission-locked PDFs are refused, because they cannot be decrypted here.
Related tools
- Flatten PDFChoose a filled-in PDF form and flatten it: every field's current value becomes fixed page content and nothing stays editable — processed on your device.
- PDF EditorOpen a PDF, add text, images, white-out boxes, highlights, shapes, drawings or a signature on any page, then save the edited PDF — all inside your browser.
- Edit PDF MetadataOpen a PDF, see its Title, Author, Subject, Keywords, Creator, Producer and dates, change any of them or clear the lot, then save a new copy — on your own device.
Prefer AllUtil on Google
One click adds AllUtil to your Google preferences. You'll see our tools highlighted with a Preferred badge in Search and AI answers.