Connect OpenClaw to Google Gemini — 1M Token Context + Free Tier

Connect OpenClaw to Google Gemini

Let me tell you about the best-kept secret in AI: Google Gemini’s free tier is absurdly generous.

While Claude and GPT charge by the token, Gemini gives you 1 million tokens of context and a solid free tier. If you’re building something that needs to process long documents or maintain extensive conversation history, Gemini is hard to beat.

Here’s how to set it up with OpenClaw.

---

Why Gemini?

Feature	Gemini	Claude	GPT-4
Free Tier	✅ Yes	❌ No	❌ No
Context Window	1M tokens	200K tokens	128K tokens
Multimodal	✅ Images, audio, video	✅ Images	✅ Images
Speed	Fast	Medium	Medium

---

Quick Start (5 Minutes)

Step 1: Get Your API Key

1. Go to Google AI Studio
2. Sign in with your Google account
3. Click “Get API Key” → “Create API Key”
4. Copy the key

Step 2: Configure OpenClaw

Option A: Environment Variable

export GEMINI_API_KEY="your-api-key-here"

Option B: CLI Configuration

openclaw config set models.providers.gemini.apiKey "your-api-key-here"

Step 3: Set Gemini as Your Model

Edit your OpenClaw config:

{
  agents: {
    defaults: {
      model: { primary: "gemini/gemini-2.5-flash" }
    }
  }
}

That’s it. You’re now using Gemini.

---

OAuth Authentication (Alternative)

For more secure, key-free authentication via your Google account:

Option 1: During Onboarding Wizard

1. Run openclaw onboard
2. Select “Google” as your AI provider
3. Choose “OAuth Authentication”
4. Complete the Google sign-in flow in your browser
5. OpenClaw stores your auth token securely

Option 2: CLI Plugin Authentication

If you’ve already installed OpenClaw, enable the Gemini CLI auth plugin:

# Enable the Gemini CLI auth plugin
openclaw plugins enable google-gemini-cli-auth

# Authenticate and set as default provider
openclaw models auth login --provider google-gemini-cli --set-default

This opens a browser window for Google OAuth. Once authenticated, you can use Gemini without managing API keys.

Note: OAuth is great for personal use. For automation or server deployments, use an API key instead.

---

Available Gemini Models

Model	Best For	Context Window
gemini-2.5-flash	Fast responses	1M tokens
gemini-2.5-pro	Complex reasoning	1M tokens
gemini-2.0-flash	Balanced	128K tokens

Model Selection

{
  agents: {
    defaults: {
      model: {
        primary: "gemini/gemini-2.5-flash",
        fallback: ["gemini/gemini-2.0-flash"]
      }
    }
  }
}

---

Understanding Gemini’s Free Tier

Google’s free tier includes:

• 60 requests per minute (Flash models)
• 2 requests per minute (Pro models)
• 1 million token context window

For most personal use cases, the free tier is more than enough.

Rate Limits

If you hit rate limits, consider:

• Switching to Flash models for higher limits
• Adding delays between requests
• Upgrading to Gemini API paid plan

---

Advanced Configuration

Full Provider Setup

{
  models: {
    providers: {
      gemini: {
        apiKey: "your-api-key",
        models: [
          {
            id: "gemini-2.5-flash",
            name: "Gemini 2.5 Flash",
            reasoning: false,
            input: ["text", "image"],
            contextWindow: 1000000,
            maxTokens: 8192
          }
        ]
      }
    }
  }
}

Using with Other Providers

Mix Gemini with other models:

{
  agents: {
    defaults: {
      model: {
        primary: "gemini/gemini-2.5-flash",  // Fast, free
        fallback: ["anthropic/claude-sonnet-4"] // For complex tasks
      }
    }
  }
}

---

Troubleshooting

”API key not valid”

Double-check your API key:

1. Go to Google AI Studio
2. Verify the key is active
3. Ensure no extra whitespace when copying

”Quota exceeded”

You’ve hit the free tier limit. Options:

• Wait for the quota to reset (daily)
• Switch to Flash models for higher limits
• Upgrade to a paid plan

”Model not found”

Verify the model ID:

openclaw models list

Make sure you’re using the correct format: gemini/model-name

Still stuck? Our AI Setup Assistant is trained on all OpenClaw documentation and can help troubleshoot your Gemini setup in real-time.

---

Gemini vs Other Providers

Aspect	Gemini	Claude	OpenAI
Best For	Long docs, multimodal	Creative, nuanced	Coding, analysis
Free Tier	✅ Generous	❌ None	❌ None
Context	1M tokens	200K tokens	128K tokens
Coding	Good	Excellent	Excellent

My recommendation: Use Gemini for document processing and as a cost-effective fallback. Keep Claude or GPT for tasks requiring the best quality output.

---

What’s Next?

• Install OpenClaw → — Full setup guide
• Use Ollama → — Free local AI
• Connect Telegram → — Control from your phone
• Docker deployment → — 24/7 availability

---

Need help? Join the OpenClaw Discord or check the official docs.