SmartSchema Extract API — JavaScript & Python Client

Convert unstructured text to validated JSON using AI. Extract structured data from any text, HTML, PDF content, or email — powered by Google Gemini Structured Output.

---

What is SmartSchema Extract API?

SmartSchema Extract API is a REST API that transforms any unstructured text into a strictly validated JSON object based on a dynamic schema you define per request.

No more brittle regex. No more manual parsing. Just send your text and your desired JSON Schema — the API returns clean, validated structured data.

Key features

• Dynamic schemas: define any JSON Schema per request, no pre-configuration needed
• Strict validation: output always matches your schema exactly
• Any text format: HTML, raw text, PDF content, email bodies, transcripts
• Zero hallucinations mode: strict_mode: true returns null instead of guessing
• Low latency: median response time under 800ms

---

🚀 Deployment Platforms

You can consume SmartSchema Extract on two different platforms depending on your needs:

1. RapidAPI (REST API + Monthly subscription)

Perfect for standard API integration where you pay a fixed monthly subscription for a guaranteed monthly quota. You don't need to manage AI keys; we handle the inference costs. 👉 Subscribe on RapidAPI

2. Apify Store (Pay-Per-Event + No-Code UI)

Ideal if you want to run extractions via a visual form, integrate with n8n/Make native nodes, or pay purely per event ($0.01 per run). Uses your own Google AI Studio API Key. 👉 Run on Apify Store

---

Quick Start (JavaScript)

1. Get your free API key

Subscribe at: rapidapi.com → SmartSchema Extract API

The Free plan includes 50 requests/month at no cost.

2. Basic call

const response = await fetch(
  'https://smartschema-extract.p.rapidapi.com/v1/extract',
  {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'X-RapidAPI-Key': 'YOUR_RAPIDAPI_KEY',
      'X-RapidAPI-Host': 'smartschema-extract.p.rapidapi.com'
    },
    body: JSON.stringify({
      text: 'Order #1042. Customer: Alice Johnson. Total: $149.99. Status: shipped.',
      schema: {
        type: 'object',
        properties: {
          order_id:      { type: 'string' },
          customer_name: { type: 'string' },
          total_amount:  { type: 'number' },
          status:        { type: 'string' }
        }
      }
    })
  }
);

const { data } = await response.json();
console.log(data);
// → { order_id: "1042", customer_name: "Alice Johnson", total_amount: 149.99, status: "shipped" }

---

Use Cases & Examples

Use case	Example file
Basic order extraction	examples/basic-extraction.js
Invoice / receipt parser	examples/invoice-parser.js
E-commerce product scraper	examples/ecommerce-scraper.js
Email entity extractor (Python)	examples/email-parser.py

---

API Reference

Field	Type	Required	Description
text	string	✅	Unstructured text to extract data from (max 100,000 chars)
schema	object	✅	JSON Schema (type: object) defining output structure
options.strict_mode	boolean	❌	Return null for missing fields instead of inferring

Pricing

Plan	Price	Requests/month	Max text
Free	$0	50	2,000 chars
Starter	$19/mo	5,000	10,000 chars
Growth	$49/mo	25,000	50,000 chars
Business	$149/mo	100,000	100,000 chars

View all plans and subscribe →

---

Error Codes

Code	HTTP	Description
MISSING_FIELD	400	text or schema not provided
INVALID_SCHEMA	400	Schema must be type: object with properties
TEXT_TOO_LONG	400	Text exceeds plan character limit
FORBIDDEN	403	Direct access without RapidAPI key
RATE_LIMIT	429	Plan request limit reached
EXTRACTION_FAILED	500	AI model error

---

Documentation

• Quick Start Guide
• JSON Schema Guide
• Use Cases

---

Contributing

Issues and PRs are welcome. See CONTRIBUTING.md.

⭐ Star this repo if you find it useful.

---

License

MIT © 2025

API on RapidAPI: rapidapi.com → SmartSchema Extract API
Actor on Apify: apify.com → SmartSchema Extract Actor