Make a PDF from a folder of images

How it works

Each file is examined and taken down the cheapest route that does not damage it: baseline JPEG is copied in untouched, PNG is stored as lossless Flate data with its alpha channel intact, and anything else is decoded once through a canvas and written as JPEG. Pages are then built in the order shown in the list. Because the routes differ, a single document can hold images that were treated three different ways, and the result panel counts how many needed the third one.

What to watch out for

Limits

Files up to about 50 MB work reliably. Beyond that, the ceiling is your device's memory rather than any rule we impose — the work happens on your own machine. Tested in Chrome 140, Firefox 143, Safari 18.

For the pile that is not all one thing

The single-format tools exist because the answer for a folder of camera JPEGs is different from the answer for a folder of screenshots. This one is for the case where you have both, plus a couple of things saved off a website, and you want one document out of the lot.

Scanned receipts for an expense claim. Photographs of a damaged parcel. A term’s worth of lecture slides saved as pictures. Pages of a passport. What these have in common is not their format — it is that they need to arrive somewhere as one file.

Three routes in, and which one your file takes

A PDF cannot hold every kind of image. It can hold two of them, and the tool prefers those wherever it can.

Baseline JPEG is copied in exactly as it arrived. No decoding, no re-compression, no quality change at all.

PNG is decompressed and written as lossless Flate image data, keeping every pixel and keeping transparency as a soft mask.

Everything else — WebP, AVIF, progressive JPEG, CMYK JPEG — is decoded once and written as JPEG at high quality. This is the only route that costs anything, and the result panel reports how many files took it, so you are never guessing.

Order is the whole point, and it is not the order they arrived in

For a single photograph the order does not matter. For twelve pages of a scanned contract it is the difference between a document and a jumble.

Operating systems hand over multiple files in an order that looks alphabetical until it isn’t: page2.jpg sorts before page10.jpg in some file pickers and after it in others, because one is comparing text and the other is comparing numbers. Relying on it is how people end up with page ten in the middle.

So the chosen files are shown as a list you can drag, and the order you leave them in is the order of the pages. Check it before converting; it takes five seconds and saves reassembling the document later.

Why a fixed page size is the default here

The single-format tools default to fitting each page to its own image, which is right when every image came off the same camera or the same screen.

A mixed pile has no single shape. Fit each page to its own image and you get a document that changes size and orientation as you scroll, which readers display at inconsistent zoom and printers handle badly. A4 with a margin is the safer default: every image is scaled down to fit inside the page and centred, nothing is cropped, and the result prints and reads as one document rather than a stack of unrelated sheets.

Nothing is uploaded

Everything happens in a Web Worker in your browser. Documents photographed with a phone are among the most sensitive files anyone converts — identity documents, signed contracts, bank letters, medical results — and none of them need to be sent to somebody else’s machine to be stacked into a PDF.

Your files never leave your browser. Last updated 2026-08-02.