Your HWP file won't open — here's why
Someone sent you a file ending in .hwp or .hwpx, and every program you own refuses to open it. Nothing is broken and the file is probably fine — it is simply a Korean document format that most software outside Korea was never built to read.
What you actually received
HWP is the document format of Hancom Office (한글), a word processor that is standard across Korean government agencies, universities, banks, and large companies. It fills the role Microsoft Word fills elsewhere, so paperwork originating in Korea — admission letters, immigration forms, contracts, public filings — routinely arrives as .hwp.
It is not a corrupted Word file and renaming it to .docx will not help. It is a genuinely different format with a different internal structure, and Word has no importer for it.
There are two variants, and knowing which one you have matters because tools differ in what they support:
- .hwp — the older format, a binary OLE2 compound file. Similar in spirit to the legacy .doc format: a container holding several internal streams.
- .hwpx — the newer format, XML documents inside a ZIP archive. Closer in spirit to .docx. If you rename a copy to .zip and open it, you will see the internal parts, including one XML file per document section.
As a rule, .hwpx is far easier for third-party software to read reliably. If whoever sent the file can re-export it as .hwpx, that alone solves a lot of compatibility problems.
Why your software refuses it
Each of the obvious candidates fails for its own reason, which is why trying them one after another feels so unproductive:
- Microsoft Word has no HWP import filter at all. It will either refuse the file or show binary garbage.
- Google Docs cannot open it either — uploading to Drive stores the file, but there is no viewer or converter for this format.
- macOS Preview and Quick Look handle PDFs, images, and a handful of office formats. HWP is not among them.
- LibreOffice and OpenOffice are the confusing case. They do ship an importer, but it only covers very old Hangul '97-era files. Documents saved by later versions changed structure enough that the old filter cannot parse them, so the import fails even though the extension looks identical. Many people conclude their file is broken when it is simply newer than the filter.
How to read it
Which route makes sense depends on whether you need to read the document or edit it.
- If you only need the content — convert it. A browser converter turns the file into PDF, Word, Markdown, or plain text without installing anything. This is the fastest path and works the same on Windows, macOS, Linux, Android, and iOS. You can do it on this page, and there are platform-specific steps for Mac and phones.
- If you need to edit it in the original format — you need Hancom Office itself. Conversion is one-directional in practice: you can get the content out, but round-tripping edits back into a pristine .hwp is not something a converter should promise.
- If the sender is reachable — ask for a PDF, or for the file re-exported as .hwpx. Thirty seconds on their side saves you the whole problem.
What to expect from a conversion
Being honest about this saves disappointment. Text and tables come through reliably. Visual fidelity depends on which variant you have.
.hwpx files saved by Hancom Office carry a typesetting cache — the layout the program had already calculated, stored inside the file. A converter can read that cache and reproduce the original page layout closely, which is why a preview of an .hwpx often looks nearly identical to the original.
Older binary .hwp files generally do not carry that information in a usable form. You will get accurate text, headings, and table structure, but the page layout is approximated rather than reproduced.
Layout reproduction also has real failure modes. In an audit of 82 real-world government .hwpx documents, we found roughly 17% rendered with content clipped off the page — in the worst case a nine-page document rendering as three. That is why our viewer detects clipping and warns you instead of silently showing an incomplete document, and why PDF export falls back to a text-based path when clipping is detected. A faithful-looking page that is missing content is worse than a plain one that has all of it.
Reading the file without uploading it anywhere
If the document is sensitive — immigration paperwork, a contract, medical or legal records — where the file goes matters as much as whether it opens.
Our converter processes files in memory and does not store them on the server. Nothing is written to disk and nothing is retained after the response is sent. If you would rather not send the file anywhere at all, asking the sender for a PDF remains the only approach that involves no third party whatsoever.