Skip to main content
Learn how to use Guardrails with Azure OpenAI Service.

​
Prerequisites

Set your Azure OpenAI credentials: 
export AZURE_API_KEY="your-azure-api-key"
export AZURE_API_BASE="https://example-endpoint.openai.azure.com"
export AZURE_API_VERSION="2023-05-15"
Or in Python: 
import os
os.environ["AZURE_API_KEY"] = "your-azure-api-key"
os.environ["AZURE_API_BASE"] = "https://example-endpoint.openai.azure.com"
os.environ["AZURE_API_VERSION"] = "2023-05-15"

​
Basic usage

from guardrails import Guard
import os

os.environ["AZURE_API_KEY"] = "your-azure-api-key"
os.environ["AZURE_API_BASE"] = "https://example-endpoint.openai.azure.com"
os.environ["AZURE_API_VERSION"] = "2023-05-15"

guard = Guard()

result = guard(
    model="azure/<<your_deployment_name>>",
    messages=[{"role": "user", "content": "How many moons does Jupiter have?"}],
)

print(result.validated_output)
Replace <<your_deployment_name>> with your actual Azure OpenAI deployment name.

​
Streaming

Stream responses from Azure OpenAI with real-time validation: 
from guardrails import Guard
import os

os.environ["AZURE_API_KEY"] = "your-azure-api-key"
os.environ["AZURE_API_BASE"] = "https://example-endpoint.openai.azure.com"
os.environ["AZURE_API_VERSION"] = "2023-05-15"

guard = Guard()

stream_chunk_generator = guard(
    messages=[{"role": "user", "content": "How many moons does Jupiter have?"}],
    model="azure/<<your_deployment_name>>",
    stream=True
)

for chunk in stream_chunk_generator:
    print(chunk.validated_output)

​
Tools/Function calling

Use Azure OpenAI’s function calling with Guardrails for structured data: 
from pydantic import BaseModel
from typing import List
from guardrails import Guard
import os

os.environ["AZURE_API_KEY"] = "your-azure-api-key"
os.environ["AZURE_API_BASE"] = "https://example-endpoint.openai.azure.com"
os.environ["AZURE_API_VERSION"] = "2023-05-15"

class Fruit(BaseModel):
    name: str
    color: str

class Basket(BaseModel):
    fruits: List[Fruit]

guard = Guard.for_pydantic(Basket)

result = guard(
    messages=[{"role": "user", "content": "Generate a basket of 5 fruits"}],
    model="azure/<<your_deployment_name>>",
    tools=guard.add_json_function_calling_tool([]),
    tool_choice="required",
)

print(result.validated_output)

​
Using validators

Add validators to ensure output quality: 
from guardrails import Guard
from guardrails.hub import ProfanityFree
import os

os.environ["AZURE_API_KEY"] = "your-azure-api-key"
os.environ["AZURE_API_BASE"] = "https://example-endpoint.openai.azure.com"
os.environ["AZURE_API_VERSION"] = "2023-05-15"

guard = Guard().use(ProfanityFree())

result = guard(
    messages=[{"role": "user", "content": "Tell me a story"}],
    model="azure/<<your_deployment_name>>",
)

print(result.validated_output)

​
Configuration

Azure OpenAI requires three environment variables:
  • AZURE_API_KEY - Your Azure OpenAI API key
  • AZURE_API_BASE - Your Azure OpenAI endpoint URL
  • AZURE_API_VERSION - The API version (e.g., “2023-05-15”)

​
Supported models

Guardrails supports all Azure OpenAI deployments including:
  • GPT-4
  • GPT-3.5-turbo
  • Any custom deployments you’ve created

​
Error handling

Guardrails automatically handles common Azure OpenAI errors with retries and exponential backoff: 
from guardrails import Guard
import os

os.environ["AZURE_API_KEY"] = "your-azure-api-key"
os.environ["AZURE_API_BASE"] = "https://example-endpoint.openai.azure.com"
os.environ["AZURE_API_VERSION"] = "2023-05-15"

guard = Guard()

try:
    result = guard(
        messages=[{"role": "user", "content": "Your prompt"}],
        model="azure/<<your_deployment_name>>",
    )
except Exception as e:
    print(f"Error: {e}")