Setting OpenAI Base URL

How to Set the OpenAI Base URL in Python, Node.js, and Environment Variables

The base_url (or baseURL) parameter tells the OpenAI SDK where to send API requests. By default, it points to https://api.openai.com/v1, but you can change it to use a relay service, self-hosted proxy, or any OpenAI-compatible endpoint. This is one of the most important configuration options for teams using API gateways or working in regions with restricted access.

Why Change the Base URL?

• API Relay/Proxy — Route requests through a service like claude4u.com for reliability, load balancing, and unified billing
• Regional Access — Access OpenAI from regions where direct connections are unreliable or blocked
• Self-hosted Gateway — Use your own API gateway for logging, rate limiting, or compliance
• Compatible APIs — Connect to any service that implements the OpenAI API interface (Azure OpenAI, vLLM, Ollama, etc.)

Python SDK Configuration

The OpenAI Python SDK accepts base_url as a constructor parameter:

from openai import OpenAI

# Method 1: Direct parameter
client = OpenAI(
    api_key="sk-your-key",
    base_url="https://claude4u.com/v1"
)

# Method 2: Using environment variable (recommended)
import os
os.environ["OPENAI_BASE_URL"] = "https://claude4u.com/v1"

client = OpenAI()  # Automatically reads OPENAI_BASE_URL

# Make a request as usual
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)

Node.js SDK Configuration

In the Node.js SDK, use the baseURL property (camelCase):

import OpenAI from 'openai';

// Method 1: Direct parameter
const client = new OpenAI({
    apiKey: 'sk-your-key',
    baseURL: 'https://claude4u.com/v1'
});

// Method 2: Environment variable
// Set OPENAI_BASE_URL=https://claude4u.com/v1 in your environment
const client2 = new OpenAI(); // Reads OPENAI_BASE_URL automatically

const response = await client.chat.completions.create({
    model: 'gpt-4o',
    messages: [{ role: 'user', content: 'Hello!' }]
});

Environment Variable Configuration

Setting the base URL via environment variables is the cleanest approach because it requires no code changes:

# Linux / macOS
export OPENAI_API_KEY="sk-your-key"
export OPENAI_BASE_URL="https://claude4u.com/v1"

# Windows Command Prompt
set OPENAI_API_KEY=sk-your-key
set OPENAI_BASE_URL=https://claude4u.com/v1

# Windows PowerShell
$env:OPENAI_API_KEY = "sk-your-key"
$env:OPENAI_BASE_URL = "https://claude4u.com/v1"

Using a .env File

For local development, store configuration in a .env file:

# .env
OPENAI_API_KEY=sk-your-key
OPENAI_BASE_URL=https://claude4u.com/v1

# Python - load with python-dotenv
from dotenv import load_dotenv
load_dotenv()

from openai import OpenAI
client = OpenAI()  # Reads from environment

// Node.js - load with dotenv
import 'dotenv/config';
import OpenAI from 'openai';

const client = new OpenAI(); // Reads from environment

cURL Configuration

When testing with cURL, simply change the URL in your request:

curl https://claude4u.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-key" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Framework-Specific Configuration

Many AI frameworks and tools support the OPENAI_BASE_URL environment variable natively:

• LangChain — Reads OPENAI_BASE_URL automatically when using ChatOpenAI
• LlamaIndex — Set api_base parameter in the OpenAI LLM constructor
• AutoGen — Configure base_url in the model config list
• Cursor IDE — Set the base URL in Settings → Models → OpenAI

Verifying Your Configuration

Run this quick test to confirm your base URL is working:

from openai import OpenAI

client = OpenAI(base_url="https://claude4u.com/v1")

models = client.models.list()
for model in models.data[:5]:
    print(model.id)

Troubleshooting

• Connection refused — Verify the URL is correct and includes the /v1 path
• SSL certificate error — Ensure the relay service uses valid HTTPS
• 404 errors — Check that the path suffix is correct (some services omit /v1)
• Environment variable not read — Restart your terminal or application after setting it