regia-ai-sdk 0.6.0

Add your description here

Regia AI SDK

A lightweight Python SDK to interact with the Regia Vision Extractor API.
Send PDF documents and automatically extract structured data using Pydantic schemas.

---

📦 Installation

Using pip:

pip install regia-ai-sdk

or uv use the command line:

uv add regia-ai-sdkß

🚀 Quickstart

from vision import VisionClient
from pydantic import BaseModel, Field
import os

class InvoiceModel(BaseModel):
  due_date: str = Field(..., description="The due date of the invoice")
  total_amount: float = Field(..., description="The total amount of the invoice")

base_url = os.getenv("BASE_URL")
token = os.getenv("API_TOKEN")

client = VisionClient(token=token, base_url=base_url)

result = client.extract("./data/sample.pdf", InvoiceModel)
print(result)

📂 Other Input Methods

From bytes

with open("./data/sample.pdf", "rb") as f:
  content_bytes = f.read()

result = client.extract(content_bytes, InvoiceModel, filename="sample_bytes.pdf")
print(result)

From base64

import base64

with open("./data/sample.pdf", "rb") as f:
  base64_string = base64.b64encode(f.read()).decode("utf-8")

result = client.extract(base64_string, InvoiceModel, filename="sample_base64.pdf")
print(result)

🧠 Features

• ✅ Simple interface for sending documents to the Regia Vision API
• ✅ Supports file path, bytes, and base64 inputs
• ✅ Use Pydantic models to define the expected data
• ✅ Automatically converts Pydantic to schema JSON

🔐 Authentication

You must provide an API token and base URL:

export BASE_URL=https://your-api-url.com
export API_TOKEN=your_api_token_here

---

Made with 💡 by Regia.ai