Document Extract
Extract structured JSON data from documents using a schema
- Define output with JSON Schema
- Type-safe structured data
- Custom extraction prompts
Document Extract
Available as an action in
Trusted by
Automate in 3 Steps
Pick Your Platform
Choose Make, n8n, or Zapier
Add PDF Vector
Search for 'PDF Vector' in apps/nodes list
Build Your Workflow
Connect to 1000+ apps
Popular Automations
Ready-to-use workflow templates for Make, n8n, and Zapier
Receipt email → Accounting
Form upload → CRM record
Invoice PDF → ERP import
For Developers
Get started in seconds. Copy and run.
import { createClient } from "@pdfvector/client";const client = createClient({ apiKey: "YOUR_API_KEY" });const result = await client.document.extract( { url: "https://example.com/annual-report.pdf", prompt: "Extract the company overview, fiscal year, revenue, net income, and key business segments", schema: { type: "object", properties: { companyName: { type: "string" }, fiscalYear: { type: "string" }, totalRevenue: { type: "number" }, netIncome: { type: "number" }, segments: { type: "array", items: { type: "object", properties: { name: { type: "string" }, revenue: { type: "number" }, growthPercent: { type: "number" }, }, }, }, }, }, model: "max", }, { context: { documentId: "my-doc-123" } },);console.log(result.data);Key Features
JSON Schema
Define your output structure with JSON Schema and get perfectly formatted data every time.
Type-Safe Output
Extracted data matches your schema types, ready for direct use in APIs and databases.
Custom Prompts
Add extraction instructions alongside your schema to guide the AI for domain-specific fields.
Credit Costs
Pick the model tier that fits your use case
Nano
Basic field extraction from simple text documents. Fastest response.
Mini
Reliable extraction from documents with tables and nested structures.
Pro
Accurate extraction from complex layouts, forms, and scanned documents.
Max
Highest accuracy for large documents with complex schemas and nested data.
Everything You Need for Document Processing
Parse, ask questions, and extract structured data from any document type
Academic Search
Search across PubMed, arXiv, OpenAlex, Semantic Scholar, and more.
Academic Fetch
Fetch paper metadata by DOI, PMID, arXiv ID, or other identifiers.
Academic Find Citations
Find relevant citations for your text using semantic matching.
Academic Paper Graph
Get a paper's citing papers and references as full publication objects.
Academic Similar Papers
Find similar papers using citation network analysis with similarity scores.
Academic Search Grants
Search grants across Grants.gov, NIH RePORTER, CORDIS, and UKRI.