Start with a non-raster PDF rewrite
Choose one unencrypted PDF, review its original size and page count, then select Preserve text & graphics and press Optimize PDF. This mode rewrites PDF objects and can pack structural overhead more efficiently without rendering pages as images. It does not intentionally lower image quality or turn selectable text into pixels. For example, a PDF with inefficiently stored objects may become smaller, while an already optimized PDF may stay the same size or grow.
The input may be up to 25 MiB and 200 pages. After writing, the tool checks that the output can be read and has the same page count, then compares actual byte sizes. A download appears only if the result is smaller. This is not a byte-identical copy: rewriting can alter metadata, form behavior, bookmarks, attachments, accessibility information, or digital signature validity. Keep the original and verify important documents.
Use image-only compression when document features are expendable
Select Image-only (lossy) only after reading its warning and checking the acknowledgment. Each page is rendered in order and embedded as a JPEG image in a new PDF. Choose 96, 120, or 150 DPI and JPEG quality of 60%, 75%, or 85%; lower settings may save more space but can blur small type. For example, a scanned image PDF can become much smaller at 120 DPI and 75% quality, but actual savings depend on the source and browser encoder.
Rasterization is not lossless. Selectable text, working links, forms, bookmarks, accessibility structure, transparency, and signatures may be lost or changed. The visible page order is kept, but page internals are replaced by images. Inspect the downloaded file at reading and print size before relying on it. If a page is too large at your chosen resolution, choose a lower DPI or use the non-raster mode.
Image-only mode accepts an input up to 10 MiB and 20 pages. It limits each rendered page to 3.5 megapixels and 2,400 pixels on either side, with 35 megapixels across the document and 25 MiB of temporary JPEG page data. These limits help prevent browser memory failures; some complex PDFs can still fail to render. A damaged or password-protected PDF is not supported.
Read the size result and keep your source
The result displays original size, actual output size, bytes saved, and percentage saved. If the output is equal or larger, it clearly says there were no file-size savings and offers no larger download. Try a lower image resolution or JPEG quality if feature loss is acceptable, or keep the original. A smaller byte count is not proof that every visual or interactive detail survived.
The selected PDF is read and processed in your browser; this tool does not upload it to a Learnvaro application backend. Browser support, memory, font rendering, and the PDF source affect results. The PDF renderer is loaded only for image-only mode. Keep an untouched backup and test the compressed file in the PDF viewer you plan to use.
PDF Compressor FAQs
Is the Preserve text & graphics mode truly lossless?
It does not rasterize pages or deliberately lower embedded image quality, so visual page content can remain unchanged. It rewrites the PDF structure, however, and does not guarantee byte-identical metadata, forms, bookmarks, attachments, accessibility data, or valid signatures.
Why did my PDF not get smaller?
A PDF may already use compact objects and efficient images. Rewriting or re-encoding can also increase bytes. The tool compares actual sizes and provides a download only when the result is smaller.
What does image-only compression remove?
It turns each page into a JPEG image. Selectable text, links, forms, accessibility structure, transparency, and digital signatures can be lost or changed. Use it only when an image-only copy is acceptable.
Which files and page counts are supported?
Use an unencrypted PDF up to 25 MiB and 200 pages for non-raster rewriting. Image-only mode is limited to 10 MiB, 20 pages, and the stated per-page and total pixel limits.
Are my PDF pages uploaded to a server?
No application upload is used for compression. The file is processed locally in this browser tab. Ordinary website requests and possible hosting or security logs are separate from processing the selected PDF.