Extract structured data from documents

Upload one or more documents together with a target JSON schema. A project is created automatically. Set auto_map to true to run the full text extraction, agent generation, mapping, and extraction pipeline in the background, or false to configure agents and the mapping later through the project endpoints.

On this page

HTTP request 

POST https://api.parsewise.ai/api/v1/extract/

Request Header

Name Required Type Description
X-API-Key Yes string API key with the pw_live_ prefix. See Authentication.

Request Body

Supported content types: multipart/form-data.

Name Required Type Description
files Yes array<string (binary)> One or more document files to process.
schema Yes any A valid JSON Schema (Draft 2020-12) describing the desired output shape.
project_name No string Optional name for the created project. Defaults to “API Extraction”.
auto_map Yes boolean When true, agents and the schema mapping are generated automatically after text extraction. When false, only the schema is stored; create agents and save a mapping with PUT /projects/{project_id}/schema/ before fetching schema-shaped results.

Responses

Status Type Description
202 V1ExtractResponse —

Security

  • ApiKeyAuth — apiKey — in X-API-Key header. API key with pw_live_ prefix.

Python example 

import os
import requests

API_KEY = os.environ["PARSEWISE_API_KEY"]
BASE_URL = "https://api.parsewise.ai/api/v1"

# Send files as multipart/form-data; repeat the "files" field per file.
with open("example.pdf", "rb") as f:
    files = {"files": f}

resp = requests.post(
    f"{BASE_URL}/extract/",
    headers={"X-API-Key": API_KEY},
    files=files,
)
resp.raise_for_status()
print(resp.json() if resp.content else None)

Definitions

V1ExtractRequestRequest

Name Required Type Description
files Yes array<string (binary)> One or more document files to process.
schema Yes any A valid JSON Schema (Draft 2020-12) describing the desired output shape.
project_name No string Optional name for the created project. Defaults to “API Extraction”.
auto_map Yes boolean When true, agents and the schema mapping are generated automatically after text extraction. When false, only the schema is stored; create agents and save a mapping with PUT /projects/{project_id}/schema/ before fetching schema-shaped results.

V1ExtractResponse

Name Required Type Description
project_id Yes string (uuid)  
status_url Yes string  
results_url Yes string