scriptiii
100% Client-Side

Rendered pages, stated trade-off

Grayscale or re-encode a PDF

Flatten a colour document to greyscale, or re-encode scans at a lower resolution to make the file smaller. Both work by rendering pages, so the trade-off is stated before you start.

Rendering happens on this device. The document never leaves the tab.

Short answer: choose the resolution, optionally keep colour, and download. This is the tool for a 12 MB phone scan of a document; it is the wrong tool for a 200 KB text PDF, which will get bigger.

Why greyscale shrinks a scan

A colour scan stores three channels per pixel; a greyscale one stores a single luminance value. For a photographed document that is roughly a third of the raw pixel data before JPEG compression, and it removes the colour cast from a phone camera at the same time. The text stays as sharp as the resolution you pick.

What you give up

The output pages are images. Links, form fields and the text layer all disappear with the original content stream, because they were never part of the picture. That is acceptable for an upload portal that only wants a smaller file, and unacceptable for a document that someone will later need to search or fill in.

Grayscale questions

Will my text still be selectable?

No, and that is the honest cost of this tool. Every page is rendered to an image and re-encoded, which is the only way to actually shrink a scan or flatten colour. If selectable text matters, use the lossless PDF compressor instead — it never re-encodes anything.

Which DPI should I choose?

150 dpi is readable on screen and prints acceptably; 200 dpi is a good middle for documents that will be printed; 300 dpi is for anything that will be OCR-ed or archived. Higher DPI means a bigger file, which works against the reason you are here.

Why is my file sometimes bigger afterwards?

A vector PDF with a few lines of text is tiny, and turning its pages into images makes it larger. The result line reports both sizes. If the result is bigger, keep the original — the tool will not pretend the conversion helped.

Does it OCR my scan?

No. It produces images, not recognised text. A scanned page that had no text layer still has none afterwards, so a “search in PDF” feature will still find nothing. OCR is a much heavier tool and is not pretended here.

Related tools

Compress PDF (lossless)Shrink without touching pages.PDF to JPGExport pages as images instead.Image converterConvert one photo instead.

Add Scriptiii to your Home Screen

  1. Tap the Share button in Safari's toolbar.
  2. Choose Add to Home Screen.
  3. Tap Add — Scriptiii opens full screen from then on.