Skip to main content
POST
/
v1
/
playground
Playground
curl --request POST \
  --url https://api.example.com/v1/playground \
  --header 'Content-Type: application/json' \
  --header 'Origin: <origin>' \
  --data '
{
  "output_schema": "<string>",
  "confidence_threshold": 123,
  "enable_verification": true
}
'
{
  "object": {},
  "metadata": {},
  "error": {}
}

​
Overview

The /v1/playground endpoint allows you to test Parsefy without an API key. It’s designed for:
  • Testing and evaluation
  • Demos and proof-of-concept
  • Learning how Parsefy works
The playground has strict rate limits. For production use, join the waitlist to get an API key and use the /v1/extract endpoint.

​
Request

​
file
file
required
The document to extract data from.
  • Supported formats: PDF (.pdf), Microsoft Word (.docx)
  • Maximum size: 10 MB
​
output_schema
string
required
A JSON Schema string defining the structure of data to extract.
​
confidence_threshold
number
default:"0.85"
Minimum confidence score (0.0 to 1.0) required before accepting Tier 1 results.
​
enable_verification
boolean
default:"false"
Enable math verification (totals, subtotals, taxes).
​
Origin
string
required
The origin of the request. Must be from an authorized domain (https://parsefy.io).

​
Response

The response format is identical to the /v1/extract endpoint:
​
object
object
The extracted data matching your schema, including _meta with confidence score and issues.
​
metadata
object
Processing information including tokens, credits, and timing.
​
error
object
Present only if extraction failed.

​
Limits

LimitValue
Request Rate1 request per second per IP
Daily Credits10 credits per day per IP
Maximum File Size10 MB
Credits reset at midnight UTC each day.

​
Example

The playground is available through the Parsefy website. Direct API calls to this endpoint require an authorized origin.

​
Response

{
  "object": {
    "merchant": "Coffee House",
    "total": 12.50,
    "date": "2024-01-15",
    "_meta": {
      "confidence_score": 0.98,
      "field_confidence": [
        { "field": "$.merchant", "score": 0.99, "reason": "Exact match", "page": 1, "text": "Coffee House" },
        { "field": "$.total", "score": 0.98, "reason": "Exact match", "page": 1, "text": "Total: $12.50" },
        { "field": "$.date", "score": 0.97, "reason": "Exact match", "page": 1, "text": "2024-01-15" }
      ],
      "issues": []
    }
  },
  "metadata": {
    "processing_time_ms": 1850,
    "credits": 1,
    "fallback_triggered": false
  }
}

​
Error Responses

​
Origin Not Allowed (403)

{
  "detail": "Origin not allowed"
}
This error occurs when the Origin header is missing or from a non-authorized domain. Solution: Use the playground through parsefy.io or join the waitlist to get an API key for direct API access.

​
Daily Limit Exceeded (429)

{
  "detail": "Daily credit limit exceeded. Resets at midnight UTC."
}
You’ve used all 10 daily credits. Solutions:
  • Wait until midnight UTC for reset
  • Join the waitlist for an API key with unlimited extractions

​
Rate Limited (429)

{
  "detail": "Rate limit exceeded. Please retry after 1 second."
}
You’re making requests too quickly. Solution: Add at least 1 second delay between requests.

​
When to Use Playground vs Extract

Use CaseEndpoint
Testing and evaluation/v1/playground
Learning the API/v1/playground
Development/debugging/v1/playground
Production applications/v1/extract
High volume processing/v1/extract
Server-side integrations/v1/extract

​
Upgrading to Production

Ready to move beyond the playground? Here’s how to switch:
1

Get an API Key

Join the waitlist to get your API key.
2

Update Endpoint

Change from /v1/playground to /v1/extract.
3

Add Authentication

Include the Authorization: Bearer pk_your_key header.
4

Remove Origin Header

The Origin header is not needed for authenticated requests.

​
Production Example

curl -X POST "https://api.parsefy.io/v1/extract" \
  -H "Authorization: Bearer pk_your_api_key" \
  -F "file=@document.pdf" \
  -F 'output_schema={...}'