Documentation

ParseJet Documentation

ParseJet extracts text from any file or URL. One API call handles PDF, DOCX, YouTube, web pages, images, audio, video, and 25+ more formats.

Quick Start

Get your first parse result in under 60 seconds. No signup required.

curl -X POST https://api.parsejet.com/v1/parse/auto/url \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com"}'

# Add your API key to requests
curl -X POST https://api.parsejet.com/v1/parse/auto/url \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com"}'

{
  "text": "Extracted text content...",
  "title": "Document Title",
  "source_type": "webpage",
  "metadata": { "url": "https://example.com" }
}

Authentication

ParseJet offers three levels of access. You can start using the API immediately without any authentication.

Core Concepts

Supported formats

ParseJet auto-detects the format from the file extension or URL pattern. You don't need to specify the format — just send the file or URL to /v1/parse/auto and ParseJet handles the rest.

Credits

Each API request consumes credits based on the complexity of the format being parsed. Simple text files cost 1 credit, while YouTube transcripts cost 5. Your monthly credit allowance depends on your plan.

Output format

By default, ParseJet returns raw extracted text. Add ?output_format=markdown to any request to get post-processed output with detected headings, lists, tables, and code blocks.

Guide

Parse a PDF

Extract text from any PDF file, including scanned documents and multi-page reports.

Upload a PDF file

curl -X POST https://api.parsejet.com/v1/parse/auto/file \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Convert to Markdown

Add output_format=markdown to preserve document structure:

curl -X POST https://api.parsejet.com/v1/parse/auto/file?output_format=markdown \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Guide

YouTube Transcripts

Get the full transcript of any YouTube video. Supports auto-generated captions in 100+ languages.

Get a transcript

curl -X POST https://api.parsejet.com/v1/parse/youtube \
  -H "Content-Type: application/json" \
  -d '{"url": "https://youtube.com/watch?v=VIDEO_ID"}'

Specify language

Use the language parameter for non-English videos:

curl -X POST https://api.parsejet.com/v1/parse/youtube \
  -H "Content-Type: application/json" \
  -d '{"url": "https://youtube.com/watch?v=VIDEO_ID", "language": "ja"}'

Or use auto-detect

The /v1/parse/auto/url endpoint automatically detects YouTube URLs:

curl -X POST https://api.parsejet.com/v1/parse/auto/url \
  -H "Content-Type: application/json" \
  -d '{"url": "https://youtu.be/VIDEO_ID"}'

Guide

Web Scraping

Extract the main content from any web page. ParseJet automatically removes navigation, ads, sidebars, and boilerplate.

curl -X POST https://api.parsejet.com/v1/parse/webpage \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/blog/article"}'

Guide

Office Documents

Parse Word (DOCX), Excel (XLSX), PowerPoint (PPTX), and CSV files. Just upload the file — ParseJet detects the format automatically.

# Works with any Office format
curl -X POST https://api.parsejet.com/v1/parse/auto/file \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

# Also works with spreadsheets
curl -X POST https://api.parsejet.com/v1/parse/auto/file \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Spreadsheets as structured JSON

Add output_format=json to get CSV and Excel rows back as a JSON array of objects — header row as keys, numeric types preserved.

curl -X POST "https://api.parsejet.com/v1/parse/auto/file?output_format=json" \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

API Reference

Response Format

All endpoints return the same JSON structure:

{
  "text": "Extracted text content...",
  "title": "Document Title",
  "source_type": "pdf",
  "metadata": { "pages": 12, "author": "Jane Doe" }
}

The recommended endpoint. Auto-detects format from file extension or URL type. Accepts file (multipart) or url (form field), not both.

curl -X POST https://api.parsejet.com/v1/parse/auto \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Parse any URL. Automatically distinguishes YouTube from regular web pages.

curl -X POST https://api.parsejet.com/v1/parse/auto/url \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com"}'

Parse any uploaded file. Detects format from file extension, falls back to content-based detection.

curl -X POST https://api.parsejet.com/v1/parse/auto/file \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Extract main content from a web page. Removes navigation, ads, and boilerplate.

curl -X POST https://api.parsejet.com/v1/parse/webpage \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/article"}'

Extract transcript from a YouTube video. Metadata includes video_id, channel, and duration.

curl -X POST https://api.parsejet.com/v1/parse/youtube \
  -H "Content-Type: application/json" \
  -d '{"url": "https://youtube.com/watch?v=VIDEO_ID", "language": "en"}'

Transcribe MP3, WAV, M4A, OGG, FLAC or WebM audio with the built-in speech recognition (runs on our servers). Max 25MB and 5 minutes of audio per request; the response carries the transcript plus timed segments. 3 credits.

curl -X POST https://api.parsejet.com/v1/parse/audio \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]" -F "language=en"

Transcribe a video: the audio track is extracted and transcribed on our servers. Supports MP4, MKV, AVI, MOV, WebM and more; up to 5 minutes of audio per request. 5 credits.

curl -X POST https://api.parsejet.com/v1/parse/video \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]" -F "language=en"

Convert a video or audio file into a downloadable audio file — MP3, M4A, Opus, WAV, or FLAC. Unlike /v1/parse/video, which returns a transcription-optimised track as JSON, this responds with the audio itself: stereo, full sample rate, and a bitrate you choose. Costs 3 credits per request.

curl -X POST https://api.parsejet.com/v1/convert/audio \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]" \
  -F "output_format=mp3" -F "bitrate=192k" \
  -o interview.mp3

Produce a file instead of text. pdf: images and PDFs are combined into one PDF in upload order (photos to PDF, merge PDFs), or a single TXT/Markdown/HTML/Office file is rendered. docx: Markdown, text, HTML, PDF, Office files, images and scanned PDFs (via OCR) become a Word document. xlsx/csv: every table found in a PDF, DOCX, HTML, CSV or XLSX. png/jpg: PDF pages rendered at a chosen resolution, or image format conversion. The response body is the file. 2 credits per request.

# PDF tables → Excel
curl -X POST https://api.parsejet.com/v1/convert/document \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]" -F "output_format=xlsx" -o invoice.xlsx

# Photos → one PDF (upload order = page order); the same call merges PDFs
curl -X POST https://api.parsejet.com/v1/convert/document \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]" -F "[email protected]" -F "output_format=pdf" -o scan.pdf

# Markdown → Word
curl -X POST https://api.parsejet.com/v1/convert/document \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]" -F "output_format=docx" -o notes.docx

Parse EPUB ebook. Extracts text organized by chapters.

curl -X POST https://api.parsejet.com/v1/parse/epub \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Parse RSS or Atom feed. Also supports OPML via /v1/parse/opml.

curl -X POST https://api.parsejet.com/v1/parse/feed \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Extract text from an image with the built-in OCR engine. Supports JPG, PNG, GIF, BMP, WebP, TIFF. Max 20MB.

curl -X POST https://api.parsejet.com/v1/parse/image \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

Same OCR as /v1/parse/image, tagged source_type "image_ocr". Scanned PDF pages are OCR'd automatically by /v1/parse/auto/file.

curl -X POST https://api.parsejet.com/v1/parse/image/ocr \
  -H "Authorization: Bearer pj_YOUR_KEY" \
  -F "[email protected]"

SDKs

Official SDKs

npm install parsejet

import { ParseJet } from "parsejet";

const client = new ParseJet({ apiKey: "pj_YOUR_KEY" });

// Parse a URL
const result = await client.parse.url("https://example.com");
console.log(result.text);

// Parse a file
const result = await client.parse.file(buffer, "report.pdf");
console.log(result.text);

// Convert a video to MP3 (3 credits)
const audio = await client.convert.audio("./interview.mov", { bitrate: "192k" });
await writeFile(audio.filename, audio.data);

pip install parsejet

from parsejet import ParseJet

client = ParseJet(api_key="pj_YOUR_KEY")

# Parse a URL
result = client.parse.url("https://example.com")
print(result.text)

# Parse a file
with open("report.pdf", "rb") as f:
    result = client.parse.file(f, "report.pdf")
    print(result.text)

# Convert a video to MP3 (3 credits)
audio = client.convert.audio("interview.mov", bitrate="192k")
audio.save()  # writes interview.mp3

AI Agents

MCP Server

Use ParseJet as an MCP (Model Context Protocol) server with Claude Code, Cursor, or any MCP-compatible AI agent.

Install

npm install -g @parsejet/mcp-server

Claude Code

Add to your project's .claude/settings.json:

{
  "mcpServers": {
    "parsejet": {
      "command": "npx",
      "args": ["-y", "@parsejet/mcp-server"],
      "env": {
        "PARSEJET_API_KEY": "pj_YOUR_KEY"
      }
    }
  }
}

Cursor

Go to Settings → MCP Servers, add a new server:

{
  "mcpServers": {
    "parsejet": {
      "command": "npx",
      "args": ["-y", "@parsejet/mcp-server"],
      "env": {
        "PARSEJET_API_KEY": "pj_YOUR_KEY"
      }
    }
  }
}

Claude.ai (Remote)

For Claude.ai web, use the remote HTTP endpoint — no local install needed:

Endpoint:  https://api.parsejet.com/mcp
Transport: Streamable HTTP
Auth:      Bearer pj_YOUR_KEY (in Authorization header)

Go to Claude.ai → Settings → Integrations → Add MCP Server → Enter the URL above.

Available tools

Rate Limits & Pricing

ParseJet uses a credit-based system. Each request consumes credits based on the format complexity.

Response headers include X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset, and Retry-After on 429 responses.

Error Codes

All errors return JSON with error and message fields.