Conversions and formats

GET /v2/info returns every conversion the API offers as one JSON array. It needs no key. The same list, grouped by input format and with a filter, is on All conversions.

Get the list

curl -sS https://api.pdfconversionapi.com/v2/info -o conversions.json
jq length conversions.json

556

The list is built into the API's gateway, so it changes only when we deploy a new version. Your code can fetch it once at startup and keep it. Three paths work without being in the list: file/to/pdf, file/to/md and office/to/pdf (below).

What an entry contains

Each entry describes one conversion. This is the entry for DOCX to PDF:

{"From":"docx","To":"pdf","Title":"DOCX to PDF","Category":"Document","Inputs":"one"}

Each entry is called as POST /v2/convert/{From}/to/{To}, for example POST /v2/convert/docx/to/pdf.

Check a conversion before you call it

jq can answer most questions about the list. This prints every target for DOCX files:

jq -r '.[] | select(.From == "docx") | .To' conversions.json | paste -sd ' ' -

docx html jpg md odt pdf png protect rtf tiff txt webp xml

And this counts the conversions whose target is PDF, then lists the conversions in the PDF Tools category:

jq '[.[] | select(.To == "pdf")] | length' conversions.json
jq -r '.[] | select(.Category == "PDF Tools") | "\(.From)/to/\(.To)"' conversions.json | paste -sd ' ' -

61
pdf/to/compress pdf/to/crop pdf/to/delete-pages pdf/to/docx pdf/to/extract pdf/to/extract-images pdf/to/flatten pdf/to/html pdf/to/image-watermark pdf/to/md pdf/to/merge pdf/to/meta pdf/to/ocr pdf/to/pdf pdf/to/pdf-watermark pdf/to/pdfa pdf/to/pptx pdf/to/protect pdf/to/repair pdf/to/rotate pdf/to/rtf pdf/to/split pdf/to/text-watermark pdf/to/unprotect pdfa/to/validate

The list holds main names only, so a lookup for jpeg finds nothing even though the API accepts it in a path. The versions below map the aliases first. Replace the paths in the loop with the ones your code calls.

# Needs the requests package: pip install requests
import requests

# Other names the API accepts in the path, and the main name each one stands for.
ALIASES = {
    "jpeg": "jpg", "jpe": "jpg", "jfif": "jpg", "tif": "tiff", "htm": "html",
    "markdown": "md", "text": "txt", "dicom": "dcm",
}


def main_name(fmt):
    fmt = fmt.strip().lower()
    return ALIASES.get(fmt, fmt)


response = requests.get("https://api.pdfconversionapi.com/v2/info", timeout=30)
response.raise_for_status()
pairs = {(c["From"], c["To"]) for c in response.json()}

for path in ["docx/to/pdf", "JPEG/to/pdf", "tif/to/png", "pdf/to/merge", "ps/to/pdf"]:
    source, _, target = path.split("/")
    supported = (main_name(source), main_name(target)) in pairs
    print(f"{path:14} {'yes' if supported else 'no'}")

// Needs Node.js 20 or later. Save it as check.mjs and run: node check.mjs
// Other names the API accepts in the path, and the main name each one stands for.
const ALIASES = {
  jpeg: "jpg", jpe: "jpg", jfif: "jpg", tif: "tiff", htm: "html",
  markdown: "md", text: "txt", dicom: "dcm",
};
const mainName = (fmt) => {
  const f = fmt.trim().toLowerCase();
  return ALIASES[f] ?? f;
};

const response = await fetch("https://api.pdfconversionapi.com/v2/info");
if (!response.ok) throw new Error(`GET /v2/info returned ${response.status}`);
const pairs = new Set((await response.json()).map((c) => `${c.From}>${c.To}`));

for (const path of ["docx/to/pdf", "JPEG/to/pdf", "tif/to/png", "pdf/to/merge", "ps/to/pdf"]) {
  const [source, , target] = path.split("/");
  const supported = pairs.has(`${mainName(source)}>${mainName(target)}`);
  console.log(`${path.padEnd(14)} ${supported ? "yes" : "no"}`);
}

Both print:

docx/to/pdf    yes
JPEG/to/pdf    yes
tif/to/png     yes
pdf/to/merge   yes
ps/to/pdf      no

We ran every example on this page on 2026-11-02 against the API server running locally in a container, with only the base URL changed (curl 8.7.1, jq 1.8.1, Python 3.14.2 with requests 2.34.2, Node.js 24.15.0).

A conversion that isn't in the list gets a 404 that names the pair, for example conversion from ps to pdf is not supported.

Format names and aliases

• Format names in the path are case-insensitive: DOCX/to/PDF and docx/to/pdf are the same conversion.
• They may contain letters, digits and hyphens, up to 20 characters. Anything else gets 400 invalid format in the URL.
• Some formats have more than one common name. The API reads each alias as the main name:

GET /v2/info and error messages always use the main name: jpeg/to/dwg gets conversion from jpg to dwg is not supported.

How many files a conversion takes

The Inputs field says which file fields a conversion reads. Requests explains the fields themselves.

Names that are tools, not formats

{to} is not always a file format. For PDFs it can name a tool: merge, split, compress, rotate, protect, unprotect, ocr, crop, flatten, repair, delete-pages, extract, extract-images, extract-attachments, meta, rasterize, resize, text-watermark, image-watermark and pdf-watermark. Our pdf/to/extract copies the pages you list in PageRange into a new PDF.

Outside PDFs:

• jpg/to/compress and png/to/compress re-encode an image to make the file smaller (PNG without loss).
• docx/to/protect, pptx/to/protect and xlsx/to/protect add a password to the file.
• png/to/qrdata, and the same path for BMP, GIF, JPG, TIFF and WebP, reads a QR code from the image and returns its text as JSON.

A few {from} names describe what you send rather than a file type:

file/to/pdf, file/to/md and office/to/pdf are not in GET /v2/info, because each one stands for many conversions. An upload whose name doesn't end in a format, such as report, gets 400 file/to/pdf needs the uploaded file's name to end in its format, such as report.docx. Options are checked as on the format's own path, so PageRange sent to office/to/pdf with a DOCX file gets PageRange is not supported yet for DOCX → PDF.

What the list does not tell you

• Which options a conversion applies. The guides for common conversions list them, for example DOCX to PDF, PDF to JPG and Merge PDF. For the rest, the rule in Requests applies: a documented option we can't apply gets a 400 that names it.
• Size and time limits. Most are the same for every conversion. Limits lists them, with the exceptions, such as HTML that needs a browser.