Compress PDF to 1MB

Pass 1 — lossless optimization. pdf-lib parses the PDF's internal object structure, strips metadata (author, editing history, embedded thumbnails), deduplicates repeated objects, and recompresses the internal streams with a tighter DEFLATE setting than most PDF exporters use by default. Nothing about how the document looks or reads changes — text stays selectable, fonts stay embedded. A typical text-heavy PDF exported from Word or Google Docs shrinks by 20-40% at this stage alone, because the bulk of an exported PDF's weight is often unused font subsets and editor metadata rather than content.

Pass 2 — page rasterization, only if Pass 1 wasn't enough. Each page is rendered to a canvas at a resolution chosen to stay legible (roughly 150-200 DPI at typical page sizes), then re-encoded as JPEG and reassembled into a new PDF. This is the same fallback technique desktop tools like Acrobat use for aggressive compression. It works on any PDF, including scanned documents that were never optimizable losslessly in the first place, but it removes the text layer — the page becomes a picture of the text rather than the text itself.

The compressor tries Pass 1 first and only escalates to Pass 2 if the result is still over your target, so a PDF that's already close to your limit may finish without losing text selectability at all.

Text PDFs exported from Word, Google Docs, or a print-to-PDF driver are usually small already, because most of the weight is fonts, metadata, and unused objects. Pass 1 lossless optimization typically brings these under 1MB while keeping the text fully selectable and searchable.

Scanned PDFs, where each page was photographed or run through a scanner, are really just images wrapped in a PDF, so they start large — a phone-scanned 10-page document at full camera resolution commonly runs 15-30MB. These need Pass 2, which recompresses every page as a high-resolution JPEG. You can reach almost any target size, but the text is not selectable unless the scan was already run through OCR before you got it.

If your PDF mixes both, such as a text report with a couple of scanned signature pages, the compressor keeps the light pages lossless where it can and only rasterizes the heavy image pages needed to reach your target — so a 20-page report with 2 scanned pages might keep 18 pages fully text-selectable.

Type any target into the size control: 1MB for most web upload forms, 2MB or 5MB for email attachments, or a strict cap like 500KB for government, visa, and job-application portals. You can enter the value in MB or KB.

The compressor works down from the original and stops as soon as the file fits, so you keep the best quality that still meets your limit. Smaller targets lean harder on Pass 2, so a scanned document squeezed to 500KB will look softer than the same file at 2MB. Need an exact image size instead? Use compress image to size.

Very long PDFs (100+ pages) that need Pass 2 rasterization can take a while and use a meaningful amount of memory, since every page becomes a full-resolution canvas bitmap before re-encoding. On memory-constrained mobile browsers this occasionally fails partway through; a desktop browser handles large documents more reliably.

PDFs with interactive form fields, embedded video, or JavaScript will have those features stripped if the file needs Pass 2 rasterization, since the output becomes a flat sequence of page images. If your PDF has fillable fields and it's already under your target after Pass 1, those fields are preserved — the loss only happens when rasterization is required.

Password-protected or encrypted PDFs need to be unlocked before compression — this tool doesn't attempt to read or crack encrypted content, it can only process a PDF it can already open.