Free · no sign-upNo watermarkNothing uploaded

Images to PDF

Put your photos on pages without anything happening to them. A JPG goes in as the bytes it already is, at the size it already is, and the tool shows you the finished file before you save it.

Nothing is uploaded, so there is no file count and no size capFull resolution, no re-encode, and the panel says so per fileAny number of pictures on any page, dragged where you want themThe exact byte count is on screen before you commit to it

Drop the pictures in, lay the pages out, then save the PDF.

No pages yet

Orientation
Margin
Picture quality

Free · no sign-up · nothing uploaded. Your pictures are turned into a PDF on your own machine, so there is no file count, no size cap and no queue. Any number of pictures on any page, dragged where you want them, and a JPG still goes in exactly as it is, at full resolution.

How to turn images into a PDF

  1. Drop your images in

    JPG, PNG, WebP, GIF and BMP, in any mixture. Nothing is uploaded, so a folder of camera photos opens as fast as a set of thumbnails and there is no file count to run into.

  2. Lay the pages out

    The strip at the bottom is the document, one group per sheet. Drag pictures between pages, or onto New page. Any page can hold any number. Then drag them around on the sheet itself and pull a corner to resize.

  3. Read the size, then save

    The rail tells you the resolution kept, whether anything was re-encoded, and the exact byte count. Then the PDF is written on your own machine.

Most converters change your picture and do not mention it

Before this tool was built, the five results ranking for "images to pdf" and "jpg to pdf" were each run on the same three files: a 3024×2016 photo, a 1700×2200 scan and a 1200×800 chart. Then the PDFs they produced were opened and read. They do not agree with each other, and not one of them tells you which kind of tool you are using.

ConverterResolutionRe-encodedSays so on the page
floiKeptNoYes, per file, before you save
SmallpdfKeptNoClaims no loss, and is right
PDF24KeptNoNo
iLovePDFCapped at 1240 pxYes, JPEG quality 60No

The last row is the site ranking first. Its output was 452,825 bytes against the 3,019,331 this tool produced from the same input, and the quality it used is written inside the file it hands back, as /DecodeParms << /Quality 60 >>. That is not a scandal. It is a reasonable default for somebody emailing a receipt, and it is roughly what you get here by choosing Smaller file. The difference is that here it is a button you pressed, with the quality, the pixel cap and the resulting byte count all named on screen. The full method and the rest of the table are in the guide.

Using the images to PDF tool

Every control on the screen, what it changes, and the four things it will refuse to do. The tool opens on an empty stage with a drop zone in the middle. Nothing below happens until there is at least one picture in it.

What you can drop in

Five formats, mixed freely in one document, in any order you like. What the PDF does with each one differs, and the Pictures row in the panel counts them by type so you can see which is which.

FormatWhat happens to it inside the PDF
JPGEmbedded as the bytes it already is. Nothing is decoded and nothing is re-encoded
PNGDecoded once and stored with Flate, which is zip. Lossless, and usually larger than the file you dropped in. Transparency is kept as a soft mask
WebP, GIF, BMPSame as PNG. PDF has no filter for these, so they are decoded and re-stored losslessly
HEICOpens in Safari, which is the only browser that will decode it. Elsewhere, convert it first

Laying the pages out

The strip along the bottom is the document, not a row of thumbnails. One bordered group per sheet, numbered, holding the pictures on it. You compose the document by dragging pictures between those groups, which is why there is no "pictures per page" setting to obey.

  • Drag a picture into another group to move it to that sheet.
  • Drag it onto New page to give it a sheet of its own.
  • Drag a whole page along the strip to reorder it. This works with a finger as well as a mouse.
  • A page with nothing left on it stops existing, rather than exporting as a blank sheet.

Where the pictures land on a shared sheet is a grid to begin with, sized to the number on that page and to which way the paper is turned. Two pictures stack on a portrait page and sit side by side on a landscape one. Then it is yours: drag any picture on the sheet to move it, and pull a corner handle to resize it.

The picture's own proportions are held whichever corner you pull, so nothing can be stretched by accident, and the opposite corner stays pinned. Pictures may overlap, and they may hang off the edge of the sheet, which is how you get a photograph that bleeds. The grid button above the preview puts a page back on its rails.

A hand-placed picture stores its centre and its width as fractions of the page, so switching from A4 to US Letter moves the layout with the paper. Rearranging the pages clears those placements: a position chosen for a sheet of two means nothing on a sheet of six.

The page controls

The bar under the preview. Every one of these applies to the whole document, and the size figure updates as you touch them.

ControlWhat it does
Page sizeFit to image sizes each page to the picture on it, so nothing is letterboxed. A4, US Letter, US Legal, A3 and A5 give a fixed sheet instead
ArrangeA shortcut, not a setting. Picking 1, 2, 3, 4, 6 or 9 writes the page breaks evenly and then gets out of the way. It reads Custom as soon as the pages stop matching it
OrientationAuto turns each sheet to suit what is on it. Portrait and Landscape force every page one way
MarginDoes double duty: it sets the page edge and the gap between pictures on a shared sheet. None, Small, Medium or Large
Number the pagesAdds a page number to the foot of every sheet
Add more, ClearAdd more appends to the document you have. Clear empties it and starts again

Picture quality, when the file has to be smaller

Original is the default and it stays the default. It embeds what you gave the tool and re-encodes nothing. Smaller file is the other setting, and it exists because a 40 MB set headed for an email attachment is a real problem.

Choosing it reveals both halves of the trade rather than hiding them. Quality is a JPEG quality slider from 30 to 95, at 75 by default. The dropdown beside it caps the long edge at 2400, 1600 or 1200 pixels, or leaves it at Full size. Both numbers, and the byte count they produce, are on screen before you save.

What the panel is telling you

Four rows, on the right, updating as you change anything. They exist because no rival prints any of them.

RowWhat it answers
PicturesHow many of each format went in. A JPG can be embedded as itself. A PNG has to be decoded and re-stored, which is lossless but larger.
ResolutionThe largest picture in the set, at the size it will be inside the PDF. If a cap is on, this row says the cap instead.
EncodingWhether anything was re-encoded, and at what quality. In the default mode it says untouched, because that is what happened.
Page size fromWhich density set the page. A scanner writes 300 dpi into the file and gets a page the size of the paper. Where nothing useful is declared, 150 dpi is assumed.

The big number above them is not an estimate. The document is genuinely rebuilt each time you touch a control, and the figure is the length of the file you are about to save. The File name box beside it sets what the download is called.

Keyboard

Four keys, and they change meaning depending on whether a picture has been hand-placed. They do nothing while you are typing in a text box.

KeyWhat it does
Previous and next page
With a hand-placed picture selected, nudge it by 0.5% of the page
Shift and an arrowNudge it by 5% of the page instead

The buttons above the preview are the keyboard route to what the strip does by dragging: move a picture earlier or later, rotate it, start a new page at it, put the page back on the grid, or remove it.

What it will not do

Four limits worth knowing before you start, rather than after.

  • A picture over 80 megapixels is refused. It has to be decoded to raw pixels to be embedded, and a file that large would take the tab down with it. The file is named and skipped rather than left to crash.
  • HEIC only opens in Safari. No other browser will decode it. Turn it into a JPG in the image converter first.
  • It builds PDFs, it does not read them. There is no merging an existing PDF, no splitting one, and no adding pages to one you already have.
  • There is no text layer. A scan goes in as a picture of a page, so nothing in it is searchable or selectable. That needs OCR, which this tool does not do.

Frequently asked questions

Is anything at all done to my pictures?

No, and it is checkable. PDF's image filter DCTDecode is JPEG, so a JPG does not have to be converted into anything: its own compressed bytes become the page. Measured on 28 August 2026 with eight camera-sized JPGs totalling 6,623,468 bytes, the PDF that came out was 6,627,520 bytes, and every image stream inside it was exactly the length of the file it came from. The 4,052 bytes of difference are the page descriptions, about 507 bytes a page. A PNG, WebP or GIF cannot be embedded as itself, because PDF has no filter for those, so it is decoded once and stored with Flate, which is zip and throws nothing away. Either way no lossy pass happens unless you press Smaller file and ask for one.

What is the biggest document it will build?

Bigger than you are likely to need, and the ceiling is your own machine rather than a rule someone wrote. The largest run in testing was 140 pages and a 51.5 MB PDF in one Chrome tab on a MacBook, with no slowdown worth reporting. Speed comes from the same place: 8 JPGs off the disk, 6.6 MB, took 547 milliseconds from dropping them in to the finished size appearing. There is nothing to upload, so a 60 MB set costs you the time to read it off your own disk and nothing else. The honest limit is memory. A very large PNG has to be decoded to raw pixels to be embedded, and anything over 80 megapixels is refused rather than left to crash the tab.

Why is my PDF as big as all the photos put together?

Because it contains all the photos. That is the correct answer for a lossless conversion and it is what you want if the PDF is going to a printer, a client or a records system. If the file has to travel through email instead, switch Picture quality to Smaller file. On the same three test files, 2.88 MB in: quality 75 at full resolution gave 1.09 MB, quality 75 capped at 1200 pixels gave 348 KB, and quality 60 capped at 1200 pixels gave 283 KB. Every one of those numbers is on screen before you save, which is the part that matters. Several popular converters apply roughly the third setting to everybody by default and tell you nothing about it.

My phone photos come out sideways in other tools. Do they here?

No. A phone does not usually rotate the picture when it takes it. It writes the camera's angle into an EXIF tag and leaves the pixels where they were, and a converter that ignores that tag gives you a document lying on its side. This one reads the tag and turns the page to match. It does that with the placement matrix in the page description rather than by redrawing anything, which is why the rotate buttons cost nothing as well. Turn a page forty times and the pixels are still the original pixels, because a rotation here is a line of maths and not a re-encode.

Will a transparent PNG stay transparent?

Yes. The alpha channel is written out as a PDF soft mask, which is the format's own way of holding opacity, so the page shows through wherever the picture is clear. Be ready for the file size. A PNG has to be decoded and re-stored, and Flate inside a PDF is a weaker compressor than a dedicated PNG optimiser, so a photographic PNG can come out several times larger than the file you dropped in. That is the price of not throwing anything away. If the size matters more than the transparency, Smaller file re-encodes to JPG, which has no transparency at all and will flatten it onto white.

Can I put more than one picture on a page?

Yes, and each page can hold a different number, which is the part no other converter here will do at all. Drag a picture into another page group in the strip to move it there, or onto New page to give it a sheet of its own. Two on the first page and three on the second is just those two drags. Nothing about the pictures changes either way: six photographs on one sheet are still six full-resolution photographs.

How does it decide where the pictures sit on a shared page?

A grid to start with, sized to the number on that sheet and to which way the paper is turned. Two pictures stack on a portrait page and sit side by side on a landscape one. Rows then close up and the block is centred, so four landscape photos on A4 sit together rather than in two clumps with a hole between them. Margin sets both the page edge and the gap between pictures.

Can I decide where each picture goes, rather than taking the grid?

Yes. Drag any picture on the page to move it, and pull a corner handle to resize it. The aspect ratio is held whichever corner you use, so a photograph cannot be stretched by accident. Pictures may overlap, and they may run off the edge of the page if you want a full-bleed photograph. The grid button above the preview puts that page back the way it was.

Can I change the order after dropping the images in?

Yes, in three ways. Drag a page along the strip at the bottom, which works with a finger as well as a mouse. Use the two move buttons above the preview, which is the keyboard route. Or drop the files in the order you want them, since the tool keeps the order it was given rather than sorting alphabetically behind your back. You can also rotate a page, delete a page, add more images at any point, and number the pages.

move openesc close