DOCX is Word in the OOXML format, with headings, tables and named styles. RTF is rich text that almost every word processor ever written can open. LibreOffice sits between the two, and this page says exactly what happens to a DOCX file on the way to becoming an RTF one.
What runs when a DOCX file becomes RTF
LibreOffice 24.2, on a machine we rent and watch. Your DOCX file is uploaded once, LibreOffice runs once, the RTF comes back, and neither file is kept. DOCX to RTF is one of the 3649 pairs that engine was probed on with a real file, which is why it has a page here and why the pairs the probe could not prove do not.
What survives from the DOCX into the RTF
An RTF you can still edit. This pair swaps the container, not the content: what comes back opens in a word processor and takes changes, which is the whole reason to do it rather than print to PDF.
Heading styles, still recognisable as headings. The RTF carries a heading level, not just larger text, which is what lets a table of contents rebuild itself and a screen reader skip around.
Tables, as tables. Rows and cells are re-described in the RTF table model rather than flattened into lines, so they stay sortable and editable.
Character formatting: bold, and the run structure that carries it.
What DOCX to RTF costs you
Exact pagination. DOCX and RTF compute line breaks and page breaks with different engines, so a document that filled four pages can come back filling five. Nothing is missing; it sits elsewhere.
Macros, and anything an application kept to itself. VBA does not cross into ODF and ODF Basic does not cross into OOXML, so a DOCX file that automates itself arrives in the RTF as one that does not.
Exotic DOCX styling with no equal on the other side. LibreOffice maps one style model onto another, and where a feature has no counterpart it lands as the nearest thing rather than as itself.
Nothing about the words themselves. This is worth separating from the paragraph above: the losses in this pair are about how the document is described, never about what it says.
Where DOCX and RTF files come from
DOCX. Word writes it, and so does Google Docs on export. It is ISO/IEC 29500, a zip of XML, which is what lets a program read the document without opening Word.
RTF. Almost everything can write it, and that is the point: an RTF is plain text you can open in a text editor and read. It was the interchange format before there was an open standard, and it survives in exports from software too old or too small to ship anything better.
If a model is going to read the RTF
Structure is what survives from the DOCX file, and structure is what a model needs. DOCX in, RTF out, with the heading tree intact instead of flattened into one long paragraph.
DOCX to RTF, measured rather than promised
Running DOCX to RTF against the real engine with a real file, LibreOffice wrote 4,205 bytes of RTF in 1.1 seconds, machine otherwise idle. That is one DOCX file on one day and not an average, which is why the number is given together with the file that produced it.
The DOCX to RTF verdict was reached by reading the internal type and finding a witness word, meaning the output was read back and recognised. Which check was used matters, because they do not all prove the same thing, and a status code of 200 proves nothing whatsoever about whether the RTF file has anything inside it.
The same DOCX file, sent somewhere else
Other outputs the probe measured out of a DOCX file, so the cost of choosing RTF can be read against something. Sizes do not compare across engines, because each family was probed with its own DOCX witness file.
DOCX to PDF: 11,917 bytes in 1.2 seconds, verified by a format signature.
DOCX to MD: 266 bytes in 0.1 seconds, verified by a pandoc round trip.
DOCX to EPUB: 5,040 bytes in 0.1 seconds, verified by a pandoc round trip.
DOCX to TXT: 264 bytes in 0.1 seconds, verified by reading the text back.
DOCX to HTML: 4,018 bytes in 0.1 seconds, verified by a pandoc round trip.
DOCX to ODT: 12,806 bytes in 1.2 seconds, verified by reading the internal type and finding a witness word.
Other ways into RTF, and what they measured
Among the published routes into RTF, DOCX is the fourth largest output of the 11 measured. The witness files differ, so this ranks the probe run and not your document.
AZW3 to RTF: 2,512 bytes in 0.6 seconds.
CSV to RTF: 779 bytes in under a tenth of a second.
HTML to RTF: 775 bytes in under a tenth of a second.
JSON to RTF: 687 bytes in 0.1 seconds.
MD to RTF: 687 bytes in under a tenth of a second.
What we will not pretend about DOCX to RTF
A DOCX file over 25 MB is refused before the upload finishes rather than after it, so you do not wait for a rejection.
A DOCX to RTF run that passes 90 seconds is killed, and the LibreOffice process is killed with it. A run left behind would sit on one of the machine's two cores until somebody noticed.
LibreOffice is serialised on purpose, one document at a time per worker, because each soffice instance takes 150 to 300 MB on a machine with 3.8 GB. Measured ceiling: 23 DOCX conversions in flight and roughly 0.9 a second sustained. Past that the queue answers 429 rather than falling over, which was tested at 32 at once.
Two thirds of the pairs you could imagine in this family do not exist, and it is worth knowing why before you go looking for one. LibreOffice converts inside whichever application opened your file: from a spreadsheet it writes spreadsheets, flat text, HTML and PDF, and it will not write a .docx no matter how it is asked. It returns nothing at all rather than something wrong. Of 113 combinations probed on this engine, 64 came back with a real file and the rest were refused outright.
This pair is one of the 64. It was proved by sending a real DOCX file through the live route and reading the RTF that came back: not its extension, its internals, plus a marker word planted in the source and found again in the output. The traits named further up this page come from a second probe that looks for each of them one by one.
One DOCX file at a time, chosen in the browser. There is nothing else to set up and nothing else on offer.
DOCX to RTF: what people ask
What actually converts my DOCX file to RTF?
LibreOffice does it, 24.2, on a machine we rent and watch. Not a browser trick and not somebody else service: the DOCX file is uploaded once, LibreOffice runs once, the RTF comes back, and neither file is kept afterwards.
How long does DOCX to RTF take?
On the file the probe used, LibreOffice took 1.1 seconds and wrote 4,205 bytes of RTF. That is one real measurement on one real DOCX file, not an average and not a promise about yours: a larger DOCX takes longer, and past 90 seconds the run is stopped.
What do I lose going from DOCX to RTF?
The one to know about first: Exact pagination. DOCX and RTF compute line breaks and page breaks with different engines, so a document that filled four pages can come back filling five. Nothing is missing; it sits elsewhere.
Can I still edit the RTF?
Yes, and that is the point of this pair rather than a PDF export. What comes back is a working RTF: open it, change it, save it. Headings are still headings and tables are still tables, so the document knows what its own parts are.
Will it look identical?
Close, not identical, and the difference is honest. DOCX and RTF lay text out with different engines, so line and page breaks can fall elsewhere and a four page document can come back at five. Nothing is missing when that happens; it sits in a different place. If you need the layout frozen exactly, convert to PDF instead, and accept that you can no longer edit it.
Is DOCX to RTF free?
There is a free allowance every month, and one DOCX to RTF conversion costs half a credit against it. When the allowance runs out the tool says so and stops, rather than quietly handing you a worse RTF.