Skip to main content

Documentation Index

Fetch the complete documentation index at: https://pdfitdown.eu/llms.txt

Use this file to discover all available pages before exploring further.

PdfItDown is designed to be flexible and highly customizable, while also providing a ready-to-use solution that you can employ in your code immediately without any modifications. Additionally, you can easily mount PdfItDown into a Starlette-based server, allowing you to go from local development to deployment in just a few lines of code! Let’s explore three levels of usage:

​
Level 1: Use PdfItDown as-is

To use the conversion features of PdfItDown, import the Converter class from the pdfconversion module: 
from pdfitdown.pdfconversion import Converter

converter = Converter()
Once you have a Converter instance, you can convert single or multiple files, as well as all files within a directory: 
converter.convert(
  file_path="business_grow.md",
  output_path="business_growth.pdf",
  title="Business Growth for Q3 in 2024"
)
converter.convert(
  file_path="logo.png",
  output_path="logo.pdf"
)
converter.convert(
  file_path="users.xlsx",
  output_path="users.pdf"
)
Each conversion operation returns the path(s) to the converted file(s).

​
Level 2: Customize

Once you are comfortable with the basics, you may want more control within your application. PdfItDown offers the perfect solution: custom callbacks for conversion. Instead of using the default callback, which can convert almost any file format to PDF, you can provide your own logic to the Converter class, focusing on specific document types.
The conversion_callback must follow a specific function signature:
def conversion_callback(
  input_file: str,
  output_file: str,
  title: str | None,
  overwrite: bool
) -> str:
  ...
While parameter names can vary, the order must remain the same.
Here are a couple of examples of what you can do: 
from pathlib import Path
from pdfitdown.pdfconversion import Converter
from markdown_pdf import MarkdownPdf, Section
from google import genai

client = genai.Client()

def conversion_callback(input_file: str, output_file: str, title: str | None = None, overwrite: bool = True):
    uploaded_file = client.files.upload(file=Path(input_file))
    response = client.models.generate_content(
        model="gemini-2.0-flash",
        contents=["Based on the attached documentation piece, please provide a summary for educational purposes that can be used as material for our developer community to grow and learn", uploaded_file],
    )
    content = response.text
    pdf = MarkdownPdf(toc_level=0)
    pdf.add_section(Section(content))
    pdf.meta["title"] = title or f"{input_file} - Summary"
    pdf.save(output_file)
    return output_file

converter = Converter(conversion_callback=conversion_callback)
converter.convert_directory(directory="docs/", recursive=True)

from pdfitdown.pdfconversion import Converter
from markdown_pdf import MarkdownPdf, Section
from google import genai
from pathlib import Path

client = genai.Client()

def conversion_callback(input_file: str, output_file: str, title: str | None = None, overwrite: bool = True):
    if Path(input_file).suffix not in [".json", ".yaml", ".yml"]:
        raise ValueError("File is not an OpenAPI spec document")
    uploaded_file = client.files.upload(file=Path(input_file))
    response = client.models.generate_content(
        model="gemini-2.0-flash",
        contents=["Can you please provide a human-readable and elegant description of the attached OpenAPI spec, with routes and associated names, paths, and request/response formats?", uploaded_file],
    )
    content = response.text
    pdf = MarkdownPdf(toc_level=0)
    pdf.add_section(Section(content))
    pdf.meta["title"] = title or f"{input_file} - OpenAPI Spec"
    pdf.save(output_file)
    return output_file

converter = Converter(conversion_callback=conversion_callback)
converter.convert(file_path="openapi.json", output_path="openapi_spec.pdf")
For more examples, check out the cookbooks in the GitHub repository!

​
Level 3: Deploy

After defining your own logic, you can easily attach the PDF converter to any Starlette or Starlette-based application (such as a FastAPI app).
To use this feature, you need to install the server variant of PdfItDown:
pip install pdfitdown[server]
Or, alternatively, install starlette:
pip install starlette
Here’s how you can mount a PdfItDown-based route for file conversion to an existing server app. Let’s start by defining an application: 
from starlette.applications import Starlette
from starlette.requests import Request
from starlette.responses import PlainTextResponse
from starlette.routing import Route

async def hello_world(request: Request) -> PlainTextResponse:
    return PlainTextResponse(content="hello world!")

routes = [
    Route(path="/hello", endpoint=hello_world, methods=["GET"], name="hello_world")
]

app = Starlette(routes=routes)
Now that the app is defined, let’s add our Converter instance: 
from pdfitdown.pdfconversion import Converter

converter = Converter()
Mounting the converter into your application is extremely simple: 
from pdfitdown.server import mount

app = mount(
    app=app,
    converter=converter,
    path="/conversions/pdf",
    name="pdfitdown",
    uploaded_file_field="file"
)
Now, run your application with uvicorn (assuming you saved it as api.py): 
uvicorn api:app --host 0.0.0.0 --port 80
The /conversions/pdf route (mounted with PdfItDown conversion features) accepts one or more files as multipart/form-data. If the request is successful, the response consists of a stream of the PDF content bytes. Here is an example of how to call the API endpoint: 
import httpx

with open("file.txt", "rb") as f:
    content = f.read()

# The field name must match the `uploaded_file_field` parameter
# used in the `mount` function
files = {"file": ("file.txt", content, "text/plain")}

with httpx.Client() as client:
    response = client.post("http://localhost:80/conversions/pdf", files=files)

    response.raise_for_status()
    with open("file.pdf", "wb") as f:
        f.write(response.content)
With this, you have completed the third and final level of PdfItDown mastery!