Skip to content

Getting Started

The Simplest Way

import strutex
from pydantic import BaseModel

class Invoice(BaseModel):
    invoice_number: str
    total: float

result = strutex.extract("invoice.pdf", model=Invoice)
print(result.invoice_number, result.total)

That's it. You're done. Everything below is optional.

Installation

pip install strutex

Optional Extras

pip install strutex[cli]          # CLI commands
pip install strutex[ocr]          # OCR support
pip install strutex[langchain]    # LangChain integration
pip install strutex[all]          # Everything

More Control: DocumentProcessor

For more control over the extraction process:

from strutex import DocumentProcessor
from pydantic import BaseModel

class Invoice(BaseModel):
    invoice_number: str
    date: str
    total: float

processor = DocumentProcessor(provider="gemini")

result = processor.process(
    file_path="invoice.pdf",
    prompt="Extract the invoice details.",
    model=Invoice
)

print(f"Invoice: {result.invoice_number}")
print(f"Total: ${result.total}")

Using Native Schemas

If you prefer not to use Pydantic:

from strutex import extract, Object, String, Number

schema = Object(properties={
    "invoice_number": String(description="The invoice ID"),
    "total": Number(description="Total amount")
})

result = extract("invoice.pdf", schema=schema)
print(result["invoice_number"])

Choosing a Provider

String providers are shortcuts for registered providers that use environment variables:

# Default: Gemini (uses GEMINI_API_KEY or GOOGLE_API_KEY env var)
result = strutex.extract("doc.pdf", model=Schema)

# OpenAI (uses OPENAI_API_KEY)
result = strutex.extract("doc.pdf", model=Schema, provider="openai")

# Anthropic (uses ANTHROPIC_API_KEY)
result = strutex.extract("doc.pdf", model=Schema, provider="anthropic")

# Local with Ollama (no API key needed)
result = strutex.extract("doc.pdf", model=Schema, provider="ollama")

For production: Use explicit provider instances for full control. See Providers.

Environment Variables

Variable Description
GEMINI_API_KEY Google Gemini API key (primary)
GOOGLE_API_KEY Google Gemini API key (fallback)
OPENAI_API_KEY OpenAI API key
ANTHROPIC_API_KEY Anthropic API key

Supported File Types

Format Extensions Notes
PDF .pdf Native support, OCR fallback
Images .png, .jpg, .tiff Vision-capable model
Excel .xlsx, .xls Converted to text
Text .txt, .csv Direct input

Next Steps