Getting Started
Parselyze turns documents into structured JSON using templates.
You define the structure you want once, then Parselyze applies it to PDFs and images through the API. OCR is used when needed, especially for scans, but the product is fundamentally template-based document extraction, not just raw OCR.
What Parselyze Does
Parselyze helps you:
- Extract structured fields from invoices, receipts, contracts, IDs, resumes, and custom documents
- Reuse the same template across many similar files
- Parse documents through a synchronous or asynchronous API
- Combine OCR + LLM-based extraction behind a stable JSON structure
How It Works
- Create a template in the dashboard Use the AI template wizard or the visual editor.
- Get your API key Generate it from your account.
- Send a document with a template ID Parselyze returns structured JSON based on that template.
Fastest Path
The easiest way to start is to upload a sample document in the dashboard and let the AI template wizard suggest the fields for you.
Processing Modes
- Synchronous API: best when you want the JSON result immediately in the response
- Asynchronous API: best for background workflows, polling, or webhook-based automation
Important Current Limits
- Templates are created in the web app, not through the public API
- The async API currently accepts one file per job
- Scanned PDFs over 30 pages are not supported yet
Next Steps
- Read the Quick Start
- Create your first template with the Template Tutorial
- Review Authentication
- Choose between Synchronous Processing and Asynchronous Processing
- Check Error Handling