Relaunching the Instaparser API
Ten years ago, we launched the Instaparser API to give developers access to the same parsing technology that powers Instapaper. Since then, the web has changed dramatically — PDFs are everywhere, LLMs need clean input, and developers need structured content more than ever.
Today, we’re relaunching Instaparser with three APIs built and refined on Instapaper:
All three are included in every plan, including the free tier which provides 1,000 API credits per month.
Article Extraction
The Article API is what powers Instapaper’s reader, and it extracts clean, structured content from any URL — title, author, date, body text, images, and metadata. Just pass a URL and get content back as structured JSON, and output the article body as html, text, or markdown.
curl "https://instaparser.com/api/1/article" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com/post"}'
{
"title": "How We Built Our Search Engine",
"author": "Jane Kim",
"date": "2026-02-15T09:00:00Z",
"content": "<p>Clean, structured HTML of the article body...</p>",
"word_count": 1847,
"thumbnail": "https://example.com/images/hero.jpg",
"excerpt": "A deep dive into building search infrastructure..."
}
What you can build with it:
- Personal article archive: Extract articles and save them as markdown files on your computer for offline reading or long-term archival.
- Content aggregation: Pull article content into newsletters, digests, or feeds without dealing with inconsistent HTML.
- RAG pipelines: Feed clean, structured web content directly into retrieval-augmented generation workflows.
PDF Parsing
Upload a PDF or pass a URL to a hosted PDF. Instaparser handles scanned documents, multi-column layouts, and tables. The API generates images for PDF figures, and output the document as html, text, or markdown.
curl "https://instaparser.com/api/1/pdf?url=https://example.com/paper.pdf" \
-H "Authorization: Bearer YOUR_API_KEY"
We’ve been building our PDF parser for Instapaper over the past couple of years, and with the Instaparser API you can now use it in your own applications.
What you can build with it:
- LLM input: Convert PDFs to clean text for use as context in large language models, without wrestling with parsing libraries or losing document structure.
- Figure and table extraction: Pull structured content from research papers or financial reports for downstream analysis or visualization.
- Document processing: Automate extraction from legal filings, academic papers, or internal documents at scale.
Summarization
The Summary API returns a concise summary of a URL with a single API call. The Summary API extracts key sentences and generates a summary from the page content — useful for previews, context compression, and content curation.
curl "https://instaparser.com/api/1/summary" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com/long-article"}'
What you can build with it:
- Link previews: Generate rich preview cards with meaningful summaries instead of truncated meta descriptions.
- Content curation: Quickly assess whether an article is worth a full read by reviewing its key sentences.
- LLM context compression: Trim long articles down to their essential points before passing them into context windows, reducing token costs.
- Email digests: Summarize a collection of articles into a digestible newsletter or daily roundup.
Getting Started
All three APIs are included in every Instaparser plan, including the free tier. Sign up for an API key and start making requests in minutes.
We offer official SDKs for Python and Node.js, or you can use the REST API directly with curl or any HTTP client.
Python
pip install instaparser
from instaparser import InstaparserClient
client = InstaparserClient("YOUR_API_KEY")
article = client.Article("https://example.com/post")
print(article.title) # "How We Built Our Search Engine"
print(article.content) # "<p>Clean, structured HTML...</p>"
Node.js
npm install instaparser-api
import { InstaparserClient } from 'instaparser-api';
const client = new InstaparserClient({ apiKey: 'YOUR_API_KEY' });
const article = await client.article({ url: 'https://example.com/post' });
console.log(article.title); // "How We Built Our Search Engine"
console.log(article.content); // "<p>Clean, structured HTML...</p>"
curl
curl "https://instaparser.com/api/1/article" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com/post"}'
No-Code & AI Integrations
Instaparser is also available through popular no-code and AI platforms:
- Claude Code: Add Instaparser as a plugin to give Claude the ability to use Instaparser to extract and summarize content during conversations.
> /plugin marketplace add instapaper/instaparser-claude-plugin > /plugin install instaparser > /reload-plugins > Instaparser extract PDF from https://bitcoin.org/bitcoin.pdf and write to markdown file - OpenClaw: Use Instaparser as a document loader to feed clean, structured web content into your LangChain pipelines.
openclaw skills install instaparser - n8n: Use the Instaparser node to extract articles, parse PDFs, and summarize content as part of your automation workflows.
- We hoped to have the Instaparser n8n node ready for launch, but it’s still in review by n8n.
- If you manage a local n8n instance, you can install the Instaparser n8n node by following these instructions.
We’re actively working on more integrations. If there’s a platform you’d like to see Instaparser on, let us know at support@instaparser.com.
Pricing
All plans include access to all three APIs. No per-feature charges or credit multipliers.
| Plan | Price | Monthly Credits | Rate Limit |
|---|---|---|---|
| Trial | Free | 1,000 | 1 req/sec |
| Beta | $150/mo | 100,000 | 5 req/sec |
| Live | $500/mo | 500,000 | 25 req/sec |
| Scale | $900/mo | 1,000,000 | 50 req/sec |
| Enterprise | Custom | 10M+ | Unlimited |
Paid plans include overage pricing so you never hit a wall mid-workflow. Full pricing details are available at instaparser.com/pricing. For Enterprise or custom volume requirements, contact us at support@instaparser.com.
Built on 15+ Years of Parsing
Instaparser isn’t a new product — it’s built on the same engine that has powered Instapaper’s reading experience since 2008, refined across billions of articles. The edge cases you’ll encounter parsing the messy, modern web, we’ve already solved.
We’re excited to open this up to developers again and can’t wait to see what you build. If you have any questions, feedback, or want to share what you’re working on, reach out to support@instaparser.com.
– Instapaper Team