Our Extract API uses advanced AI to analyze PDF documents and extract structured data based on your JSON Schema. Send us a PDF URL or file along with a prompt describing what to extract and a JSON Schema defining the structure, and we'll return perfectly formatted data that matches your schema.

JSON Schema is a standard way to define the structure of JSON data. By providing a schema, you ensure the extracted data always follows your expected format - with correct types, required fields, and validation. This makes it easy to integrate extracted data directly into your applications without additional parsing or validation.

Common uses include extracting data from invoices, receipts, contracts, forms, reports, research papers, and financial statements. Any PDF with structured information can be processed - from simple forms to complex multi-page documents with tables and nested data.

Pricing is based on document pages - 3 credits per page. This transparent model ensures you only pay for what you use. The same rate applies whether extracting simple or complex data structures.

Our AI achieves high accuracy by understanding document context and structure. It handles variations in formatting, missing fields, and complex layouts. The schema validation ensures extracted data meets your requirements, and you can make prompts more specific to improve accuracy for your use case.

Yes! The API supports complex JSON Schemas including nested objects, arrays, conditional fields, and custom validation rules. You can extract hierarchical data, tables with multiple rows, related entities, and any structure that can be defined with JSON Schema.

Document files must not exceed 200 MB and should be no longer than 1,000 pages. JSON Schemas should be reasonable in size and complexity. The API works best with text-based PDFs rather than scanned images.