Skip to main content

​
Prerequisites

Before you begin, you’ll need:
  • API Key - See Authentication for how to obtain your API key
  • HTTP Client - Any tool that supports multipart/form-data requests (curl, Postman, etc.)
  • Sample Documents - PDF, JPG, PNG, HEIC, or TIFF files for testing

​
Your First API Call

Test the API with a health check: 
curl https://api-vision.fintelite.ai/health
Response: 
{
  "status": "healthy",
  "timestamp": "2024-01-15T10:30:00Z"
}
Most endpoints require authentication. See Authentication for details on using your API key.

​
Basic Workflow

Extract structured data from your document by uploading it directly: 
curl -X POST https://api-vision.fintelite.ai/predict-async \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "files": "https://example.com/invoice.pdf",
    "schema": {
      "type": "object",
      "properties": {
        "invoice_number": {
          "type": "string"
        }
      }
    }
  }'
Use async endpoints (/predict-async, /parse-async, /fraud-async) to avoid timeouts. Synchronous endpoints (/predict, /parse, /fraud) are not recommended for production use.
For complete request/response examples and all available parameters, see the Predict Async API Reference.

​
Using Templates

For repeated extractions, create reusable templates with a custom ID: 
curl -X POST https://api-vision.fintelite.ai/template \
  -H "X-API-Key: YOUR_API_KEY" \
  -F 'name=Invoice Extractor' \
  -F 'custom_id=INVOICE' \
  -F 'description=Extract invoice data' \
  -F 'schema={"type":"object","properties":{"invoice_number":{"type":"string"}}}'
Then use the template by its custom ID: 
curl -X POST https://api-vision.fintelite.ai/predict-async \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "files": "https://example.com/invoice.pdf",
    "template_id": "INVOICE"
  }'
Using custom_id makes your template references readable and memorable instead of random UUIDs. You can also use the auto-generated UUID if you prefer.
For template management details, see the Templates Concept Guide.

​
Checking Job Status

Check the status of your async job: 
curl https://api-vision.fintelite.ai/status/job-uuid \
  -H "X-API-Key: YOUR_API_KEY"
For webhook notifications and complete job status details, see the Jobs Concept Guide.
For comprehensive error codes and handling, see the Error Codes section in API Overview.

​
Best Practices

Use File IDs

Upload once, reference by ID for multiple operations to save bandwidth

Batch Processing

Use async mode for processing multiple documents efficiently

Template Reuse

Create templates for recurring document types to maintain consistency

Monitor Confidence

Check confidence scores to identify low-quality extractions

​
Next Steps

Document Processing

Learn about processing modes and options

Templates

Master template creation and management

Fraud Detection

Add document authenticity verification

API Reference

Explore all available endpoints