[bug] Bug Report: 404 Error Using Guardrails Server with Azure OpenAI API

[OrenRachmil]

Describe the bug
I am trying to use the Guardrails AI server framework with the Azure OpenAI API to validate LLM output. Despite modifying the example to fit Azure OpenAI, I encounter a 404 - {'detail': 'Not Found'} error. Below are the details of my setup, code, and debugging steps.

I started the Guardrails server with the following command:
guardrails start --config config.py
I used the following Python code to connect to the Guardrails server and proxy my Azure OpenAI API request:

from openai import AzureOpenAI
# Set up Azure OpenAI client with Guardrails server acting as the endpoint
client = AzureOpenAI(
    azure_endpoint="http://localhost:8000/guards/gibberish_guard/openai/v1",
    api_key='api-key',
    api_version="2024-09-01-preview"
)

# Send request through the Guardrails proxy
response = client.chat.completions.create(
    model="model name",
    messages=[{
        "role": "user",
        "content": "Make up some gibberish for me please!"
    }]
)

# Access the validated response
print(response.choices[0].message.content)
print(response.guardrails['validation_passed'])

When running this code, I get the following error:
openai.NotFoundError: Error code: 404 - {'detail': 'Not Found'}

Request for Assistance

1. Is there an example of using Guardrails with the Azure OpenAI API, or specific adjustments required to make them compatible?
2. How can I configure Guardrails to handle Azure OpenAI properly with custom validation routes?
3. I would also like to integrate the server with litellm to access other models beyond OpenAI. Is there documentation or guidance for such integration?