How-to2026-10-034 min read

Convert a PDF to JPG or PNG: What DPI Changes and How to Export One Page

PDF to Images renders every page into a ZIP, as PNG unless you pick JPEG. The browser form has only format and DPI; API fields for page selection and WebP are ignored. Measured file sizes at 150 and 300 DPI, and how to export a single page.

PDF123 Β· Updated 2026-10-03

PDF to Images turns every page of a PDF into one image file and returns them in a ZIP. The browser form has two settings, format (PNG or JPEG) and DPI, and there is no way to pick pages or choose WebP. Both settings change the output size a lot, and the sections below show by how much.

A worked example: six pages, four settings

The test file was a 6-page, US Letter, text-only document generated with Ghostscript: 32 lines of 11 pt Times per page, 5,593 bytes. It was sent to the /api/v1/convert/pdf/img endpoint of a local build of the server, once per setting. The numbers are from one run on 2026-10-04; another document will give different sizes.

DPI Format Pixels per page Page 1 ZIP, all 6 pages
150 PNG 1275 x 1650 53,400 bytes 195,476 bytes
300 PNG 2550 x 3300 127,391 bytes 494,897 bytes
150 JPEG 1275 x 1650 252,280 bytes 1,340,684 bytes
300 JPEG 2550 x 3300 685,891 bytes 3,272,299 bytes

Pixel dimensions follow from the page size: a Letter page is 8.5 by 11 inches, so 150 DPI is 1275 by 1650 and 300 DPI is 2550 by 3300. Doubling DPI quadruples the pixel count, but the PNG ZIP grew about 2.5 times here, because flat white areas compress well.

For this file, JPEG was larger than PNG: page 1 was about 4.7 times the size at 150 DPI. The usual advice that JPEG is the small format holds for photographs. A text page is mostly flat white with sharp edges, which PNG compresses well and JPEG does not. A photographic or scanned page may go the other way, so test one representative page before you convert a hundred.

The ZIP is named after the PDF (contract.zip for contract.pdf) and contains page.1.png through page.6.png, or .jpg files when you choose JPEG.

Options the API accepts but ignores

The browser form does not offer these two, but a script that sends them gets no error and no effect:

  • Page selection. Sending pageNumbers=2 returned all six pages, not one. Every page is always rendered.
  • WebP. imageFormat=webp returned PNG files with the same names and sizes as png. Only jpeg and jpg select JPEG; any other value falls back to PNG.

If you need WebP, convert the PNG files afterward in an image editor.

Getting a single page

Cut the page out first, then convert the one-page PDF:

  1. Open Extract Pages, upload the PDF, and enter 2 as the page to keep. The result in the test was a 1-page PDF.
  2. Upload that PDF to PDF to Images, set DPI to 150, and keep PNG. The ZIP contained one file, page.1.png.

The file is named page.1.png even though it came from page 2 of the original, because numbering restarts in the extracted PDF. Rename it if the original page number matters.

DPI: where the form and the API disagree

The web form pre-fills 300. A direct API call that omits dpi renders at 150. If a script produces smaller images than the browser did, this is why; send dpi explicitly in both. A non-numeric value is rejected with a 400 and the message "dpi must be a number".

Rough guidance, grounded in the table above and not in any quality benchmark: 150 DPI is enough to read a full page on a screen; 300 DPI is for print or for zooming into fine print. In this run the 300 DPI PNG ZIP was about 2.5 times the 150 DPI one, and a long scan-heavy document multiplies that.

What does not survive

The result is pixels, so:

  • Text is no longer selectable or searchable. For text, use PDF to Text, or OCR for a scan.
  • Links stop working.
  • Vector logos and diagrams are rasterized at the DPI you chose and will blur if enlarged later. Rerun at a higher DPI instead.

Keep the original PDF next to the images. Going back with Images to PDF gives you a PDF made of pictures of text, not the original document.

A password-protected PDF is rejected with a 400, This PDF is password-protected. Enter its password., and a password field sent to this endpoint does not change that. Remove the password with Unlock first.

Check before you bulk-use the set

Open one page from the ZIP at 100% zoom and check small text and thin lines. If it is soft, raise DPI and rerun; if the ZIP is too large, lower DPI, or switch to JPEG only after confirming it is actually smaller for your pages.

Open tool
Process in the browser β€” no watermark, files removed after the job.
Open tool