Gretel announces strategic Azure integration to accelerate privacy-preserving AI development

Large language models (LLMs) have revolutionized how organizations approach AI development, but accessing and using sensitive data for model customization remains a significant challenge. Today, we're excited to announce Gretel's integration with Microsoft Azure AI Foundry Model Catalog. Azure Teams can now use Gretel’s  compound AI system, Navigator directly from their Azure environment and start generating high quality synthetic data that enables secure  and privacy-preserving AI training. This integration represents more than just a technical partnership - it's about removing critical data  bottlenecks  that have historically forced organizations to choose between innovation and privacy, and providing a path to scaling AI responsibly.

With Navigator now on the Azure AI Foundry Model Catalog, customers get:

• Enterprise-Grade Infrastructure: Seamless scalability, availability, and reliability of Navigator through Azure’s infrastructure.
• Streamlined Workflow: Simplified deployment and management and reduced operational overhead while generating synthetic data.
• Cost-Effective Synthetic Data Generation: Pay-as-you-go pricing and elimination of manual data collection costs.
• Enhanced Data Quality: Consistent, well-structured data generation with built-in quality validation.

Use Cases

We’ve used Navigator to create interesting open datasets for fine-tuning code generation models on Text-to-SQL or Text-to-Python datasets, align models to generate non-harmful content, and many other essential enterprise applications.  You can use Navigator to create datasets specific to your use case that can  augment and enhance your model’s performance. 

Getting Started is Simple

Customers can use a simple prompt to iteratively create, edit, and expand high-quality tabular datasets. 

1. Find Navigator on AI Foundry Model Catalog

1.1 Find Navigator in your Model Catalog

1.2 Deploy Gretel-Navigator-Tabular

Once you deploy the model, you will see your Endpoint URI and API key. 

2. Import the Gretel SDK in your Notebook

Copy the endpoint and API key and use them in the following code snippet.

pip install -U -qq openai gretel-client
from openai import OpenAI
from getpass import getpass
from gretel_client import Gretel

# Set region and get credentials securely
AZURE_ENDPOINT = <YOUR_ENDPOINT_URI>
AZURE_API_KEY = getpass("Azure API key: ")
oai_client = OpenAI(base_url=AZURE_ENDPOINT, api_key=AZURE_API_KEY)
azure_open_ai = Gretel.create_navigator_azure_oai_adapter(oai_client)

3. Generate your data

import pandas as pd

PROMPT = """Generate a mock dataset for users from the Foo company based in France.
Each user should have the following columns:
* first_name: traditional French first names.
* last_name: traditional French surnames.
* email: formatted as the first letter of their first name followed by their last name @foo.io (e.g., jdupont@foo.io)
* gender: Male/Female
* city: a city in France
* country: always 'France'.
"""

table_headers = ["first_name", "last_name", "email", "gender", "city", "country"]
table_data = [
    {
        "first_name": "Lea",
        "last_name": "Martin",
        "email": "lmartin@foo.io",
        "gender": "Female",
        "city": "Lyon",
        "country": "France",
    }
]

SAMPLE_DATA = pd.DataFrame(table_data, columns=table_headers)

metadata, synthetic_df = azure_open_ai.generate(
    "gretelai/auto",
    PROMPT,
    num_records=10,
    sample_data=SAMPLE_DATA,
)

print(synthetic_df)
print("*******")
print(metadata)

You can find a getting started notebook on GitHub so you can try this out for yourself!

Join Our Growing Synthetic Data Community

Become part of the thriving Gretel ecosystem and connect with fellow innovators: 

• Engage with Experts:  join us and other developers, engineers, data scientists, and privacy fans in the Synthetic Data Community on Discord to discuss synthetic data, share experiences, and get direct support from our team
• Stay Informed: Follow us on LinkedIn, YouTube, and X for the latest updates on synthetic data innovations and privacy-preserving AI best practices. ‍
• Get Exclusive Updates: Subscribe to our newsletter for early access to product announcements, technical insights, and success stories from organizations leveraging Gretel on Azure

Ready to transform your AI development workflow? Contact our team today to learn more about how Gretel and Azure together can accelerate your AI initiatives while ensuring data security, privacy and compliance.