Skip to main content
OpenAI Codex and the OpenAI CLI use OPENAI_BASE_URL to determine the API endpoint. Set this to https://api.tokios.com along with your Tokios key, and Codex routes to your local model through the Tokios tunnel.

​
Prerequisites

​
Set up the environment

codex-windows.cmd
set OPENAI_BASE_URL=https://api.tokios.com
set OPENAI_API_KEY=sk-tokios-YOUR_KEY_HERE
codex --model gemma-tunnel
Replace gemma-tunnel with the exact model name you registered in the Tokios dashboard.

​
Using the OpenAI Python SDK

If you prefer to call your local model programmatically, initialize the OpenAI client with the Tokios base URL and your API key:
openai_tokios.py
import openai

client = openai.OpenAI(
    base_url="https://api.tokios.com/v1",
    api_key="sk-tokios-YOUR_KEY_HERE",
)

response = client.chat.completions.create(
    model="gemma-tunnel",
    messages=[{"role": "user", "content": "Explain async/await in Python"}],
)
print(response.choices[0].message.content)
The OpenAI Python SDK requires the base URL to end with /v1. The Tokios endpoint is https://api.tokios.com/v1.