Quickstart
Generate your first synthetic dataset in under 5 minutes using the Python SDK, CLI, or REST API.
5 min read · Updated 12 Mar 2026
Overview
01
Sign up & get your API key
Create a free account at hadarac.com. Your API key is under Settings → API Keys. Free plan includes 15 generations/month.
Get your API key → 02
Install the SDK or CLI
Run pip install hadarac in any Python 3.8+ environment. The CLI and Python SDK are the same package — both installed together.
View on PyPI → 03
Generate your first dataset
Write one sentence describing your schema. Hadarac uses Gemini 2.5 to produce rows that match your description — including realistic correlations.
04
Export & integrate
Download as Parquet, CSV, or JSON. Load directly into pandas, dbt, or any pipeline tool. Pull programmatically via the REST API.
View API reference →Code examples
# 1. Install the SDK
pip install hadarac
# 2. Set your API key
import os
import (hadarac)
client = (hadarac).Client(api_key=os.environ["HADARAC_API_KEY"])
# 3. Generate 500 rows of e-commerce orders
dataset = client.generate(
prompt="E-commerce orders: order_id, product_name, quantity, price, country",
rows=500,
format="parquet",
)
# 4. Save locally
dataset.save("orders.parquet")
✓ Generated 500 rows in 1.4s → orders.parquetNext steps
← Previous
Authentication