Google (Gemini)
Gemini is Google's family of AI models. The standout advantage of Gemini is its generous free tier — you can get started without a credit card, making it the easiest provider to try.
Step-by-Step Setup
1. Sign In to Google AI Studio
- Go to aistudio.google.com.
- Sign in with any Google account (the same one you use for Gmail, YouTube, etc.).
- No credit card is needed. No phone verification is needed.
Gemini is the only major AI provider where you can go from zero to a working API key in under two minutes with no payment information at all.
2. Accept Terms of Service
On your first visit, Google AI Studio will ask you to accept the Terms of Service. This is a one-time step.
3. Create an API Key
- In the left panel of Google AI Studio, click "Get API key".
- Click "Create API key".
- You may be asked to select an existing Google Cloud project or create a new one. If you do not have one, AI Studio will create one for you automatically.
- Copy the API key — it starts with
AIzaSy.
Unlike some other providers, Google lets you view your existing API keys at any time from the AI Studio dashboard. But it is still good practice to copy and save your key right away.
4. Connect to OpenClaw Host
- Open your OpenClaw Dashboard.
- Go to Instance > Integrations tab.
- Find the Gemini CLI card and click Connect.
- Paste your API key into the field.
- Click Save.
Your instance will now use Google Gemini as its AI provider.
Free Tier Limits
The free tier is available to everyone, no credit card required:
| Model | Requests per Minute | Requests per Day | Tokens per Minute |
|---|---|---|---|
| Gemini 2.5 Pro | 5 | 100 | 250,000 |
| Gemini 2.5 Flash | 10 | 250 | 250,000 |
| Gemini 2.0 Flash-Lite | 15 | 1,000 | 100,000 |
For personal use and experimentation, the free tier is often sufficient. Gemini 2.5 Flash at 250 requests per day covers a good amount of coding agent usage. If you hit limits, you can upgrade to pay-as-you-go through Google Cloud.
AI Studio vs Google Cloud
Google offers AI access through two surfaces:
| Google AI Studio | Google Cloud (Vertex AI) | |
|---|---|---|
| Cost | Free tier available | Pay-as-you-go only |
| Credit card | Not required | Required |
| Best for | Personal projects, exploration | Production workloads, enterprise |
| API key type | Simple API key | Service account or OAuth |
| Rate limits | Lower (see table above) | Higher, configurable |
For OpenClaw Host, Google AI Studio keys work perfectly. You only need Vertex AI if you are running a large-scale production deployment and need higher rate limits.
Optional: Gemini CLI
If you want to use Gemini directly on your local machine (outside of OpenClaw Host), you can use the Gemini CLI:
Installation
Quick try (no install needed):
npx @google/gemini-cli
This downloads and runs Gemini CLI temporarily without a global install. Great for trying it out.
Global install (npm, requires Node.js 18+):
npm install -g @google/gemini-cli
Windows: If you do not have Node.js installed:
winget install OpenJS.NodeJS.LTS
Then open a new terminal and run:
npm install -g @google/gemini-cli
Authentication
Gemini CLI supports three authentication methods:
Option A — Google OAuth (recommended):
gemini
On first run, it opens a browser-based Google login. This method gives you higher rate limits than an API key because it uses your Google account's own quota.
Option B — API key:
export GEMINI_API_KEY="AIzaSy-your-key-here"
Uses the free-tier limits shown in the table above.
Option C — Vertex AI (enterprise):
export GOOGLE_CLOUD_PROJECT="your-project-id"
export GOOGLE_APPLICATION_CREDENTIALS="/path/to/service-account.json"
For organizations using Google Cloud with service accounts.
You do not need to install Gemini CLI locally to use it through OpenClaw Host. The CLI section is here for users who also want Gemini on their own machine.