How to obtain a Gemini API key — Google AI Studio vs. Kunavo

A Gemini API key is a bearer credential that authenticates calls to Google's Gemini models. There are two ways to get one: get an official key from Google AI Studio, or use Kunavo to call Gemini, Claude, GPT, and image/video models with one OpenAI-compatible key (no Google Cloud project required). This guide compares the two options and shows the fastest way to start making calls in production.

Method 1 — Google AI Studio (official)

If you plan to call only Gemini, getting a key directly from Google is a good fit.

1. Visit aistudio.google.com and sign in with your Google account.
2. Create a key via Get API key → Create API key. The key has the format AIza....
3. The free tier for prototyping has rate limits and is suitable for early development.
4. To handle production volume, you need to link the key to a Google Cloud billing account with billing enabled.

This is the right path if Gemini will remain the only model you call. But if you also use Claude or image/video models, you'll manage separate keys, SDKs, and bills for each provider.

Method 2 — Kunavo (recommended for production)

Kunavo puts Gemini behind an OpenAI-compatible endpoint. Use one sk-kn- key to call all models and pay from a single Stripe balance, so no Google Cloud project is needed. Gemini is around 70% cheaper than Google's list price; your balance never expires, and it's pay-as-you-go with a $10 minimum top-up.

1. Sign up — create an account at kunavo.com.
2. Add $10 — pay as you go, and your balance never expires.
3. Create a key — create a key at /app/keys. It starts with sk-kn- and is shown only once, so save it in an environment variable.
4. Make a call — set the OpenAI SDK's base_url to https://api.kunavo.com/v1 and call a Gemini slug.

Quick start

Use the official OpenAI SDK as is. Just change base_url and the model name — nothing else needs to change.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["KUNAVO_API_KEY"],
    base_url="https://api.kunavo.com/v1",  # 바뀌는 건 이 줄뿐입니다
)

# 모델 슬러그만 바꾸면 Claude, GPT 등 다른 모델도 같은 키로 호출됩니다
resp = client.chat.completions.create(
    model="gemini-2-5-flash",
    messages=[{"role": "user", "content": "두 문장으로 자기소개를 해줘."}],
)
print(resp.choices[0].message.content)

Keep your key secure

• Use environment variables. Don't hard-code the key in your code; store it in an environment variable such as KUNAVO_API_KEY. Browser and mobile bundles are public, so never expose the key on the client; always proxy through your own server.
• Rotate it regularly. Kunavo allows multiple active keys, so you can rotate them with a clean overlap rather than a hard cutover. Quarterly rotation is recommended.
• Set spending limits per key. Set a spending limit for each key in the dashboard and check per-key usage at /app/usage, so if a key is leaked, you can revoke it without affecting the others.

Frequently asked questions