parserbee-python

Python client for the ParserBee API. Send a PDF, scan or phone photo of an invoice, purchase order, receipt or bank statement, and get back structured JSON that matches your Parser Template.

Install

pip install parserbee

Requires Python 3.9 or later.

Before you start

1. Create a free account at app.parserbee.com. New accounts get 25 free pages, no credit card.
2. Create a Parser Template: upload one sample document and the AI builds the fields for you, or pick a ready-made one from the template library.
3. Copy your API key from Settings (it starts with pk_) and the template ID from the template page.

export PARSERBEE_API_KEY=pk_your_api_key

Quick start

from parserbee import ParserBee

client = ParserBee()  # reads PARSERBEE_API_KEY

result = client.extract("your_template_id", "invoice.pdf")
print(result["data"])
# {'invoice_number': 'INV-2026-001', 'vendor_name': 'Acme Corp', 'total_amount': 1500.0, ...}

Extract from a public link instead of a local file:

result = client.extract("your_template_id", file_url="https://example.com/invoice.pdf")

An open binary file works too:

with open("receipt.jpg", "rb") as f:
    result = client.extract("your_template_id", f)

The response

extract() returns the API response as a dict:

Each extraction costs 1 page credit per page processed, with a minimum of 1.

Other methods

client.list_templates()          # every Parser Template in the account
client.get_template("tpl_id")    # one template with its fields
client.get_credits()             # balance, plan and subscription status
client.list_activity()           # the last 10 extractions
client.get_activity("req_id")    # one extraction, with its data

Errors

Every error is a subclass of parserbee.ParserBeeError, with .code and .status:

from parserbee import InsufficientCreditsError

try:
    client.extract("your_template_id", "invoice.pdf")
except InsufficientCreditsError:
    print("Top up page credits in the dashboard")

Examples

The examples folder has scripts you can run as they are:

• extract_invoice.py: one file, printed as JSON
• extract_from_url.py: a document that is already online
• folder_to_csv.py: every PDF and image in a folder, one CSV row each
• check_credits.py: balance and template IDs

Supported files

PDF, PNG, JPG, WebP and HEIC, up to 50 MB. The API also reads DOC, PPT, PPTX, XLSX, CSV, TXT, RTF, ODT, HTML and EML; see the API reference for details. Scans and phone photos work, and there is no need to run OCR first. DOCX is not accepted: save it as PDF first.

No code needed?

If you only need the data in a spreadsheet, you do not need this library. Data Lab in the dashboard takes a batch of documents and exports one combined Excel, CSV or Google Sheets file. To try a single invoice in the browser, use the free invoice parser.

Choosing a tool for invoices? Our comparison of invoice OCR software covers 10 options, tested on real invoices, scans and phone photos.

Links

• ParserBee
• API reference
• Pricing
• Support: support@parserbee.com

License

MIT