ArticleJune 19, 2026

The 5-Minute Migration: Swapping Your OpenAI Base URL to OpenDunes

Migrating your application's AI gateway doesn't require rebuilding your architecture. Swapping base URLs and API keys takes minutes.

Published: June 19, 2026
Reading: 2 min read
Code in: Python · TypeScript · Env · JSON

When you hear about switching your application's infrastructure to a local, Law 18-07 compliant gateway, you probably expect the worst: weeks of code refactoring, learning a new proprietary SDK, dealing with undocumented endpoints, and fixing integration bugs.

For developers and engineering leads, code stability is paramount. You cannot afford to halt your development pipeline to restructure your AI integrations.

Fortunately, you do not have to.

OpenDunes is built on Bifrost OSS, a high-performance proxy architecture that maintains absolute parity with the standard OpenAI and OpenRouter API specifications. To migrate your application, swap your payment headaches for Dinars, and secure compliance, you only need to change two variables in your codebase.

Here is a step-by-step developer guide to configuring the OpenDunes API proxy (openai api proxy).

The Drop-In Replacement Mechanics

The OpenDunes gateway mirrors the standard OpenAI request and response JSON schemas exactly. When your backend dispatches a payload to our endpoints, we handle auth verification, DZD wallet balance checks, and compliance filters before routing the request upstream. The response is returned to your app in the exact format your existing code expects.

To migrate, you only need to modify:

The Base URL Endpoint: Route requests to our local gateway.
The API Authorization Key: Replace your OpenAI bearer token with your Dinar-funded OpenDunes key.

1. Python Integration Example

If your backend is built on Django, FastAPI, or Flask, here is how to swap the configuration using the official openai Python library:

import os
import openai

# Swapping the base client configuration
client = openai.OpenAI(
    base_url="https://opendunes.com/v1",  # Local gateway endpoint
    api_key=os.environ.get("OPENDUNES_API_KEY")  # Your od-xxxxx key
)

# Your query logic remains 100% unchanged
response = client.chat.completions.create(
    model="gpt-4o",  # Standard model names are supported directly
    messages=[
        {"role": "system", "content": "You are a helpful database admin."},
        {"role": "user", "content": "Draft a SQL query to fetch daily active users."}
    ]
)

print(response.choices[0].message.content)

import OpenAI from 'openai';

const openai = new OpenAI({
  baseURL: 'https://opendunes.com/v1', // Swapped URL
  apiKey: process.env.OPENDUNES_API_KEY,  // Your local DZD key
});

async function generateResponse() {
  const completion = await openai.chat.completions.create({
    messages: [{ role: 'user', content: 'Generate CSS grid container queries.' }],
    model: 'gpt-4o',
  });

  console.log(completion.choices[0].message.content);
}

# OpenDunes API Configuration
OPENAI_API_BASE=https://opendunes.com/v1
OPENAI_API_KEY=od-your-dinar-funded-api-key-here

{
  "error": {
    "message": "Insufficient balance in OpenDunes wallet. Please fund your account in DZD.",
    "type": "insufficient_balance",
    "code": "insufficient_credits"
  }
}

The 5-Minute Migration: Swapping Your OpenAI Base URL to OpenDunes | OpenDunes

The 5-Minute Migration: Swapping Your OpenAI Base URL to OpenDunes

The Drop-In Replacement Mechanics

1. Python Integration Example

2. Node.js / TypeScript Example

3. Environment Variable Configuration

Wallet Management & Error Handling

Build in Dinars, Query Globally

Frequently Asked Questions

Does the proxy add latency to my API queries?

Do I need to change my model parameter calls?

How do I track token usage for billing?

Rest of the series

Related articles