How to convert images to PDF without uploading them
Every "free" online images-to-PDF converter you've used uploads your photos to a company server, converts them there, and — in several documented cases — keeps a copy or a live download link for days. For personal photos, scans of IDs, and anything you wouldn't post publicly, that's an unnecessary risk. The safer approach is to convert the file where it already is: on your own device.
This tool does exactly that. When you drop images here, your browser:
- Checks each file is really the image it claims to be — by its magic bytes, not its name — and that the file is complete, so a truncated download is refused with a clear message instead of producing a PDF with a broken image. An image saved with the wrong extension still converts, and a renamed impostor (a PNG wearing a .jpg name) is refused with a pointer to the right converter.
- Reads each photo's EXIF orientation data and stores the output upright, so a portrait phone photo doesn't come back sideways. The rest of the metadata, including location, is dropped.
- Embeds JPGs byte-for-byte and rasterizes the rest with pdf-lib — JPGs go in with no decoding and no re-encoding, so there is no quality loss at all; HEIC, AVIF, GIF, TIFF, SVG, WebP, BMP and 16-bit or animated PNGs are drawn to JPEG on a canvas at the quality you choose. Your original files are never modified.
What makes this converter different
The pages that top the results for this search mostly convert on a server (iLovePDF, Smallpdf, Adobe and Convertio all upload your files first), and the free tiers cap you — one caps free conversions at 5 MB per file. The client-side pages we found accept only JPG, PNG and WebP, or only a handful of formats: none of them accepts HEIC, AVIF, TIFF or SVG at all, and none mixes all ten formats in one batch. This page is the only dedicated images-to-PDF page we found that combines a batch of 20 mixed-format images with a ZIP download, a choice of one combined PDF or one PDF per image, Original/A4/Letter page sizes, fit or fill, JPEG quality control, per-image rotation and reordering, and a live network-request counter so you can verify the privacy claim yourself. It runs entirely in your browser — there is no server to upload to.
When is images-to-PDF the right choice?
Whenever the destination wants a document, not a photo: job portals and application forms that only accept PDFs, visa and government forms, university and school uploads, scanned receipts and certificates emailed as one file, and anything that must print exactly as it looks. A PDF also keeps a set of photos in one file, in order, where a folder of images would arrive scattered. If the destination accepts images, keep the originals — a PDF adds page overhead and no quality.
Honest limitations
No text layer
The PDF pages hold pictures, not text. The PDF is not searchable, and you cannot select or copy words from it. Run it through OCR afterwards if you need searchable text.
Rasterized formats are pictures, not vectors
HEIC, AVIF, GIF, TIFF, SVG, WebP and BMP become JPEG pictures inside the PDF. An SVG's vector shapes do not stay vector, and no format gains a text layer. The JPEG quality setting is the only quality control, from 50 to 95.
First frame only for animations
Animated GIFs, animated WebP and animated SVGs contribute their first frame only, and a multi-image HEIC contributes its first still only. The PDF format has no animation, so later frames are dropped, never silently merged.
8-bit RGB output
Every page is 8-bit RGB. 16-bit PNGs and TIFFs are downsampled to 8-bit, transparency is flattened onto the background colour you choose, and CMYK TIFFs are converted to RGB. Nothing is gained in the process — the PDF is exactly as good as the images you put in.
No password or encryption
This tool does not add a password or encrypt the PDF, and it does not accept password-protected or DRM-protected inputs. If a form needs a protected PDF, add the password afterwards in a desktop PDF tool.
Metadata is stripped
Camera metadata (EXIF, including location) is dropped when the image is embedded. The orientation tag is read and applied first, so the output is stored upright.