Summarize PDFs with Python — PDF Summary API Tutorial

Prerequisites

• Python 3.8+ and the requests library (pip install requests).
• An API key: create a free account and copy your key from the dashboard. Store it in an environment variable rather than in code:

export PDF_SUMMARIZE_KEY="sk_live_..."

Your first PDF summary in Python

import os
import requests

API_URL = "https://pdf-summarize.com/api/summarize.php"
API_KEY = os.environ["PDF_SUMMARIZE_KEY"]

def summarize_pdf(path, length="medium", fmt="bullets", lang="auto"):
    with open(path, "rb") as f:
        resp = requests.post(
            API_URL,
            headers={"Authorization": f"Bearer {API_KEY}"},
            files={"file": (os.path.basename(path), f, "application/pdf")},
            data={"length": length, "format": fmt, "outputLang": lang, "includeKeywords": "1"},
            timeout=180,
        )
    data = resp.json()
    if resp.status_code != 200 or not data.get("success"):
        raise RuntimeError(f"{resp.status_code}: {data.get('error', data)}")
    return data

result = summarize_pdf("annual-report.pdf")
print(result["summary"])
print("Pages:", result["page_count"], "| Keywords:", result.get("keywords"))

The response

{
  "success": true,
  "summary": "- Revenue grew 6% to ...",
  "keywords": "revenue, margin, guidance, ...",
  "page_count": 48,
  "original_length": 182340,
  "summary_length": 1210,
  "processing_time": 6.214,
  "requests_remaining": 4987
}

requests_remaining is your remaining monthly quota (-1 means unlimited).

Streaming the summary

For long documents or interactive apps, stream the summary as it's generated:

import json

def stream_summary(path):
    with open(path, "rb") as f:
        with requests.post(
            API_URL + "?stream=1",
            headers={"Authorization": f"Bearer {API_KEY}"},
            files={"file": (os.path.basename(path), f, "application/pdf")},
            data={"length": "long", "format": "mixed"},
            stream=True, timeout=300,
        ) as resp:
            resp.raise_for_status()
            for line in resp.iter_lines(decode_unicode=True):
                if not line or not line.startswith("data: "):
                    continue
                payload = line[6:]
                if payload == "[DONE]":
                    break
                event = json.loads(payload)
                if "error" in event:
                    raise RuntimeError(event["error"])
                if event.get("type") == "summary":
                    print(event["content"], end="", flush=True)

stream_summary("research-paper.pdf")

Processing a folder of PDFs

import pathlib, time

out_dir = pathlib.Path("summaries")
out_dir.mkdir(exist_ok=True)

for pdf in sorted(pathlib.Path("pdfs").glob("*.pdf")):
    target = out_dir / (pdf.stem + ".md")
    if target.exists():
        continue                      # resume-friendly
    for attempt in range(3):
        try:
            data = summarize_pdf(str(pdf), length="long", fmt="mixed")
            target.write_text(data["summary"], encoding="utf-8")
            print("ok", pdf.name, data["page_count"], "pages")
            break
        except RuntimeError as e:
            if str(e).startswith("429"):
                time.sleep(30 * (attempt + 1))   # rate limit: back off
            else:
                print("failed", pdf.name, e)
                break

Errors to handle

Tips

• Set timeouts generously (2-3 minutes) for long PDFs.
• Cache results by file hash to avoid paying twice for the same document.
• Keep keys server-side — never ship them in a front-end or mobile app.
• Full parameter reference: the API documentation.

Frequently asked questions