PNG to PDF
Built in your browser · nothing is uploaded
Turning images into a PDF wraps each one in a page of the size you choose and scales it to fit inside the margin. JPEGs are embedded exactly as they are, with no re-encoding and no quality loss. PNG and WebP are written as lossless pixel data, with any transparency preserved as a soft mask.
How to make a PDF from images
Two different things happen depending on what you drop in, and the difference decides the size of the file you get back.
A JPEG is copied into the document byte for byte and written with the DCTDecode filter, which is the PDF format's own name for JPEG data. Nothing is decoded, nothing is re-encoded, and the photo in the PDF is bit for bit the photo you started with. Anything else, PNG and WebP included, has to be decoded to pixels first, and those pixels are stored losslessly and compressed. Lossless is the right choice for a screenshot or a logo, where re-encoding would smear the text, and it is expensive for a large photograph: a twelve-megapixel PNG is around 36 million bytes of raw colour before compression, so a PDF built from PNGs can be many times the size of one built from the same pictures as JPEGs.
Transparency survives. A PNG with an alpha channel gets its transparency written as a soft mask alongside the image, so whatever is behind it on the page shows through rather than turning into a black or white rectangle.
Fit the image is the option most people want and few people find. Instead of placing the picture on a standard sheet, it makes each page exactly the size of the image plus the margin, which is what a screenshot or a scan needs; a fixed A4 page around a wide screenshot leaves two large white borders and shrinks the content. It converts at 96 pixels to the inch, the CSS convention, so a 1920-pixel-wide screenshot becomes a page 20 inches across. That is fine on screen and absurd on paper, so choose a real paper size whenever the result is going to be printed.
One small detail saves renaming files afterwards: a single image gives a PDF named after that image, and several give images.pdf.
What people use it for
- Turning phone photos of a signed document into one file
- Meeting a form that accepts a PDF and refuses a JPEG
- Putting screenshots in a fixed order for a bug report
- Making a print-ready A4 sheet out of a single image
- Combining scanned receipts into one expense claim
- Wrapping a wide screenshot in a page that is actually its shape
Questions
JPEGs are copied into the document byte for byte, so they are untouched. Everything else is stored as lossless pixels, which can make the PDF large for big images.
Because they were not JPEGs. PNG and WebP are decoded and written as lossless pixel data, which is far larger than a photographic compression would be.
For photographs, yes, and the file will be much smaller. For screenshots, diagrams and anything with text, no: JPEG artefacts around sharp edges are worse than the size.
It is kept. A PNG with a transparent background gets a soft mask in the PDF, so whatever is behind it shows through.
Yes. Leave the orientation on Auto and each page follows the shape of its own image, based on whether it is wider than it is tall.
It sizes each page to the image plus the margin instead of using a standard sheet, so there are no white borders. It is the right choice for screenshots and scans.
It converts at 96 pixels to the inch, the CSS convention. A 1920-pixel screenshot therefore becomes a 20-inch page, which is fine on screen and wrong for printing.
Scaled to fit inside the margin on all four sides and centred, keeping its proportions. Nothing is cropped or stretched.
There is no fixed limit. Each one is held in memory until you build the PDF, so a long run of large photographs is what will slow a phone down first.
Yes, with the arrows next to each thumbnail, and you can remove any you did not mean to include. Pages come out in the order shown.
Only if your browser can decode HEIC, which Safari can and most others cannot. Convert them to JPEG first and they will embed without re-encoding.
One image gives a PDF named after that image. Several give images.pdf.
No. This wraps images and nothing else, which is what keeps the output predictable and the file small.
No. The PDF is assembled in this page and handed straight to your downloads folder.