Turn PDF pages into real vector SVG files. Text arrives as text you can select and retype, lines and shapes arrive as editable paths, and the fonts are written into the file so it looks the same everywhere. It runs on your device — the PDF is never uploaded.
Quick answer: Drop a PDF below, optionally set a page range, and click Convert to SVG. Each page becomes its own .svg file with true vector paths and live text — not a picture in an SVG wrapper. Download the pages one by one or as a ZIP. Nothing is uploaded.
Most browser-based PDF-to-SVG converters render the page to a bitmap and wrap that bitmap in an <svg> element. The file has the right extension and scales without visible pixels in a browser, but nothing inside it is editable: no selectable text, no separate shapes, and it looks soft the moment you zoom past the resolution it was rendered at.
This tool reads the PDF's own drawing operators and writes each one as its SVG counterpart. A line of body copy becomes a <text> element with its original glyph positions. A logo's outline becomes a <path>. A filled table cell becomes a rectangle with the fill colour it had in the PDF. Open the result in Inkscape and you get the objects, not a photograph of them.
The result list after conversion labels every page vector or image so you always know which you got, and the counts underneath tell you how many paths and text runs came through.
Fonts are the part everyone gets wrong
An SVG that names a font it does not carry will render in whatever the viewer happens to have. Line lengths change, text overflows its boxes, and a careful layout falls apart on someone else's machine.
With Embed fonts on, every font the page uses is written into the SVG as a base64 @font-face rule, including the substitutes for the standard PDF fonts. Nothing is fetched when the file is opened, so it survives being emailed, dropped into a repo, or opened offline. It costs roughly 30–150 KB per file. Switch it off when you are producing many pages for a site that already loads those fonts.
Where it will not help
A scanned document has no vector data in it. The page is a photograph, and no converter can invent outlines that were never stored — you will get the scan embedded in an SVG, correctly placed, and the page will be labelled as an image. To get editable outlines out of a scan you need tracing, which is a different job: export the page with PDF to JPG, then run it through Vectorize.
A handful of PDF features have no SVG equivalent — some gradient meshes, certain transparency groups, a few unusual colour spaces. Those pages fall back to an embedded rendering rather than producing something broken, and the page list says so.
Frequently Asked Questions
A real vector. The PDF's drawing instructions are translated into SVG elements one for one — text becomes <text>, lines and shapes become <path>, and fills keep their colours. Nothing is rasterised, so you can zoom to any size and edit individual objects in Illustrator or Inkscape. The one exception is artwork that was already a photo or scan inside the PDF: a picture cannot become a path, so it stays a picture.
Yes. Each text run is an SVG <text> element with the original glyph positions, so you can select it in a browser, search it, and retype it in a vector editor. This is the main reason to choose SVG over PNG or JPG, which flatten everything to pixels.
Leave it on unless you have a reason not to. With it on, each font used on the page is written into the SVG as a base64 @font-face rule, so the file renders identically on a machine that has never seen that font — it is genuinely self-contained. The cost is size: an embedded font typically adds 30–150 KB per file. Turn it off for a much smaller SVG that falls back to whatever similar font the viewer has installed.
Yes. SVG has no concept of pages, so a 10-page PDF produces 10 SVG files. Convert the whole document or enter a range like 1-3, 7. When there is more than one page you also get a ZIP of the full set alongside the individual download links.
You get an SVG containing the scan as an embedded image, correctly placed and sized. That is the honest result — a scan holds no vector data to recover. If you want editable outlines from a scan, run it through an image tracer instead: export the page as PNG with PDF to JPG, then trace it on the Vectorize tool.
A few PDF features — certain gradient meshes, some transparency groups, and a handful of exotic colour spaces — have no direct SVG equivalent. Rather than dropping them or producing a broken file, that page is rendered and embedded as an image, and the result list tells you which pages that happened to. Every other page is still true vector.
No. The PDF is parsed and the SVG is written entirely in your browser, so a contract, drawing or statement never leaves your device. There is no account, no queue and no size limit beyond your own memory.
Yes — it is a standard .svg file. Inkscape and Illustrator open it with the paths and text as separate, editable objects. Figma imports it as a frame of vector layers. Because images and fonts are written in as data rather than links, nothing goes missing when you move the file.