Google

Google AI Studio provides Gemini models. The gateway reaches them through Google’s OpenAI-compatible Chat Completions endpoint: send requests to /v1/chat/completions with an OpenAI client and change only the model name. The native Gemini generateContent API and the OpenAI Responses API (/v1/responses) do not reach Google models.

Google is a built-in provider with ngrok.ai inference available, so a provider key is optional. Add your own Google key to bill usage to your Google account instead of your ngrok credits.

Setup

Create an access key

Follow the quickstart to create an access key in app.ngrok.ai.

Store your Google API key (optional)

Skip this step to use ngrok.ai inference. To bring your own key, generate one at aistudio.google.com, then add a provider key.

Create a configuration

Create an access key configuration with a google routing rule and allow Google in the access scope.

Assign and send requests

Assign the configuration to your access key and send requests with your access key:

from openai import OpenAI

client = OpenAI(
    base_url="https://gateway.ngrok.ai/v1",
    api_key="ng-xxxxx-g1-xxxxx",
)

response = client.chat.completions.create(
    model="gemini-3.6-flash",
    messages=[{"role": "user", "content": "Hello!"}]
)

Available models

See the model catalog.

gemini-2.5-pro, gemini-2.5-flash, and gemini-2.5-flash-lite need a Google project created before mid-2026. Newer projects get a 404 stating the model is no longer available to new users, and the ngrok-managed Google project postdates that cutoff. Reach these models with your own key on an older project, or use a Gemini 3 model instead.

Next steps

⌘I

Concepts

Getting Started With

SDKs

Guides

Observability

Examples

Reference

API Reference

Setup

Available models

Next steps

​Setup

​Available models

​Next steps

Setup

Available models

Next steps