PDF를 PDF/A로

PDF/A-2b, the default, suits most archiving and accepts transparency and JPEG 2000 images. Pick PDF/A-1b only when a recipient demands it, since it's the strictest level and forbids transparency and layers, and PDF/A-3b when the archive needs to carry embedded files, such as the XML data used by e-invoicing formats.

Ghostscript rewrites the PDF so it meets the standard: every font is embedded as a subset, colors are converted to device-independent color with an sRGB output intent, and features the chosen level forbids are dropped rather than failing the whole run. Page content stays intact and text remains selectable.

PDF/A requires all fonts to be embedded, so any font the original referenced but didn't include is now stored inside the file, and the sRGB color profile is added too. Compress PDF can reduce the size; re-validate afterward if strict compliance matters.

Only if the normal conversion fails. Pre-flatten renders every page to a 300 DPI image before Ghostscript runs, which gets around broken fonts or unusual transparency but leaves you with no selectable text, so try without it first. It also needs the PyMuPDF module, which loads on first use.

Not guaranteed. Ghostscript produces conformant output for the vast majority of files, but it isn't a validator, so if a court, repository, or auditor requires proof, check the result with a dedicated PDF/A validator such as veraPDF.

Yes. Add multiple files and they're converted one after another with the same settings; a single file downloads directly and several download together as pdfa-converted.zip.

No. Ghostscript runs as a WebAssembly module inside your browser, so the conversion happens on your machine and the files never leave it. The module itself is fetched from a CDN the first time you use the tool, which is why the first run takes a little longer.

The tool asks for the open password first, then converts the decrypted file. PDF/A itself forbids encryption, so the output is always unprotected; a password-protected PDF/A isn't possible.