Convert files to PDF and back: every route, explained
A PDF converter turns other file formats into PDF, or turns PDF pages back into editable formats such as Word, Excel, or images. There is no single converter that handles every format well, because some conversions can run locally in a browser while others need a full layout engine. This page maps each direction — into PDF and out of it — to the specific tool or method that does it reliably.
"Convert anything" tools quietly upload your file to a server running Word, LibreOffice, or an OCR engine, because browsers cannot render those formats themselves. This site only ships converters that genuinely run on your device, and routes everything else to an honest step-by-step method instead.
How to pick the right PDF conversion route (step by step)
- Identify your direction and format. Decide whether you are converting to PDF or from PDF, and note the exact source format, since JPG, DOCX, and EPUB each need a different route.
- Use an in-browser tool for renderable formats. Images, plain text, Markdown, and SVG render natively in a browser, so the matching converter on this site produces the PDF locally with no upload.
- Use the owning application for office formats. Word, Excel, PowerPoint, and OpenDocument files need a full layout engine, so export to PDF from the application itself or from LibreOffice, following the format's guide page.
Converting into PDF, by source format
Images. This is the easy direction: browsers decode images natively, so the conversion runs entirely on your device. Use JPG to PDF for photos, PNG to PDF for screenshots and graphics with transparency, WEBP to PDF for web-saved images, and SVG to PDF for vector artwork. Several images of mixed types can go into one document with Images to PDF. iPhone photos in HEIC and multi-page TIFF scans need a decode step first — the HEIC to PDF and TIFF to PDF guides cover it.
Text-based formats. Plain text and Markdown have simple, well-defined layout rules, so Text to PDF and Markdown to PDF paginate them locally in the browser. Web pages are best captured with the browser's own print engine — the Webpage to PDF guide shows the settings that matter.
Office documents. Word, PowerPoint, and Excel files store content plus styling rules, not final layout; something must compute every line break, page break, and cell width, and only a word processor's engine does that faithfully. Follow the guides for Word, DOCX, PowerPoint, and Excel — including the legacy .doc, .ppt, and .xls variants — or the OpenDocument routes for ODT and ODS. E-books (EPUB to PDF) and RTF files sit in the same category.
Converting out of PDF, by target format
To images. Rendering PDF pages to pictures runs fully in-browser here: PDF to JPG for compact shareable images, PDF to PNG for lossless captures, or the general PDF to Images exporter. If you want the photos embedded inside a PDF at original resolution rather than page snapshots, use Extract PDF images.
To text and web. PDF to Text pulls all selectable text into a .txt file locally, and PDF to HTML produces a simple web page from it. Both only see text that exists as text — a scanned page yields nothing without OCR, which no in-browser tool here performs.
To editable documents. This is the hardest direction. A PDF records where each glyph sits on the page, not what paragraph, table, or heading it belonged to, so recovering an editable file means reconstructing structure that was thrown away. The guides for PDF to Word, PDF to DOCX, PDF to Excel, and PDF to PowerPoint explain which tools reconstruct best and what still breaks. Specialized targets — SVG, EPUB, and the archival PDF/A standard — each have their own page.
How it works: why PDF conversion is asymmetric
PDF is a final-form format. Defined in ISO 32000, it describes pages as fixed drawing instructions: place this glyph at these coordinates, paint this image in this rectangle. That design is why conversion into PDF is fundamentally a printing problem — something renders the source and the result is captured as page descriptions — while conversion out of PDF is a reconstruction problem, closer to reverse engineering than translation.
Going in, fidelity depends entirely on who does the rendering. The application that owns the format applies its own line-breaking, font substitution, and pagination rules, which is why Word's own PDF export beats any third-party converter, and why a browser can flawlessly convert the formats it already knows how to draw — images, text, SVG — but nothing else. When the original application is unavailable, LibreOffice's engine is the best free approximation for office formats, and the guide pages above give its exact commands.
Coming out, the converter reads positioned glyphs and must infer everything else: that these glyphs form words, the words a line, the lines a paragraph, the aligned runs a table column. Heuristics do this well for simple single-column documents and progressively worse for multi-column layouts, floated figures, and dense tables. Scanned PDFs are the extreme case — their pages are photographs containing no text objects at all, so text-direction converters return nothing until an OCR engine has added a recognized text layer.
This asymmetry yields a practical rule: always keep the editable source file, and treat PDF as the distribution copy. Converting out of PDF is a recovery operation for when the source is lost — not a routine round trip.
Worked example: one report, three conversions
A 12-page grant report exists as report.docx (86 KB). Exporting it from Word or LibreOffice produces report.pdf at 412 KB — bigger because fonts are now embedded, but every heading, link, and page break lands exactly as designed. Later, the funder asks for the summary chart as a standalone image: PDF to PNG renders page 3 at 1654 × 2339 pixels (150 DPI equivalent) into a 690 KB PNG, entirely in the browser.
Months later the .docx is lost and the report needs updating. A PDF-to-Word reconstruction of the 412 KB file recovers roughly 95% of it cleanly — body text, headings, and the simple tables — but the two-column references section comes back as one interleaved column that takes ten minutes to repair by hand. Total round trip: seconds into PDF, minutes of cleanup out of it, which is the asymmetry in miniature.
PDF conversion — frequently asked questions
What is the best free way to convert any file to PDF?
Print to PDF is the universal fallback: every application that can print can produce a PDF through the system print dialog. Format-specific exporters give better results — bookmarks, links, and selectable text — so use those when the application offers one.
Why can some conversions run in the browser while others cannot?
Browsers can already decode images, text, and SVG, so wrapping those into a PDF works locally with JavaScript. Word, Excel, and EPUB files describe their layout abstractly and need a full rendering engine to decide line breaks and page breaks, which browsers do not expose.
Does converting a PDF to Word give me back the original document?
No. A PDF stores positioned glyphs, not paragraphs and styles, so the converter must guess the structure, and complex layouts come back imperfect. Scanned PDFs contain no text at all and need OCR first.
Is converting an image to PDF lossless?
For JPG it can be fully lossless, because the JPEG bytes are embedded in the PDF unchanged. PNG and WEBP are decoded and re-wrapped but keep every pixel value, so nothing visible is lost either way.
Which conversions on this site happen without uploading my file?
All live tools here — images to PDF, PDF to images, text and Markdown to PDF, PDF to text and HTML — run entirely in your browser. Guide pages cover the formats that need desktop software, and those files never touch this site at all.