Images to PDF takes any number of JPEG, PNG, or WebP files and assembles them into one PDF document. The conversion runs entirely in your browser using pdf-lib, an open-source JavaScript library that constructs the PDF byte stream directly. Files are loaded into a Blob in tab memory, rendered onto PDF pages, and serialized to a download — there is no upload step and no server-side processing.
Each image becomes its own PDF page. You can choose a fixed page size (A4 at 595×842 points, US Letter at 612×792, or Legal at 612×1008) or pick auto-fit, which sizes each page to match the image's pixel dimensions converted at 72 DPI. Margins are configurable per side, and the fitting mode controls whether the image stretches to the box, fits inside it preserving aspect ratio, or fills the box with cropping.
JPEGs are embedded with their original DCT-compressed bytes intact, so there is no quality loss for photographs. PNGs and WebPs are decoded and re-encoded as PDF image streams; pdf-lib uses Flate (zlib) compression for these, which preserves every pixel but yields larger files than JPEG would. If file size matters, convert PNG screenshots to JPEG first — the savings are significant for non-transparent content.
The drag-and-drop list lets you reorder images before generating; the export follows that order. Individual images can be removed without re-uploading the rest. There is no batch rotation control inside this tool — if a scan is sideways, rotate it in the source first or run the resulting PDF through the PDF Page Organizer afterward.
Browser memory is the practical limit. Around 100–150 high-resolution photos (each 4–8 MB) is comfortable on a modern laptop; beyond that, encoding all pages in one pass can pressure the tab's heap and cause Chrome to terminate it. Splitting the job into batches and merging the resulting PDFs with the PDF Merger is the standard workaround.
The output is a 1.7-spec PDF that opens in macOS Preview, Adobe Acrobat, Chrome's built-in viewer, and any PDF/A-compliant reader. It is not flagged as image-only or scanned, so subsequent OCR will work normally. If you need a searchable text layer, run the result through the PDF OCR tool.
For a single image (e.g., a passport scan that a portal demands as PDF), this tool still works and produces a one-page file in seconds. For long-form documents originally written in Word or Google Docs, prefer the Word to PDF tool — it preserves text as text rather than rasterizing each page.