Google Gemini

Google’s Gemini models offer strong performance with unique features like Google Search grounding, extended thinking, and built-in image generation.

Setup

1. Get an API Key

1. Go to aistudio.google.com
2. Sign in with your Google account
3. Click “Get API Key”
4. Create a new key or copy existing

2. Configure in BraceKit

1. Open Settings → AI Provider
2. Click Gemini in the provider grid
3. Paste your API key
4. Select a model from the dropdown (models are fetched automatically)

Settings are saved automatically as you type.

Available Models

Note: Models are fetched dynamically from Google’s API based on your API key.

Current Stable Models

Image Generation Models

Thinking Models

Gemini 2.5 Pro supports extended thinking for complex reasoning:

• Click the brain icon (🧠) in the toolbar to enable
• The model shows its reasoning process before the response

Note: Image generation models don’t support extended thinking or Google Search.

Special Features

Google Search Grounding

Gemini can use Google Search to provide up-to-date information:

1. Click the Google icon in the toolbar (when using Gemini)
2. Send your message
3. Gemini searches and includes sources

You: What are the latest features in React 19?

BraceKit: [Uses Google Search]

Based on recent information, React 19 includes:

1. **Actions** — New way to handle form submissions
2. **use() hook** — Read resources in render
3. **Server Components** — Improved support

Sources:
- react.dev/blog
- github.com/facebook/react

Note: Google Search grounding is not available on image generation models.

Image Generation

Generate images directly in chat:

1. Select gemini-2.5-flash-image model
2. Choose an aspect ratio in the toolbar
3. Describe the image you want
4. The image appears in the response

See Image Generation for full details.

Vision

All Gemini models support image analysis:

• Attach images to messages
• Ask questions about them
• Get detailed analysis

Large Context

Gemini models support large context windows (up to 1M tokens):

• Analyze entire documents
• Process large codebases
• Extended conversations

Model Parameters

Configure in Settings → AI Provider under the Advanced section:

Aspect Ratios for Image Generation

When using gemini-2.5-flash-image:

Note: Gemini doesn’t support “auto” aspect ratio. Select a specific ratio.

Deprecated Models

The following models are deprecated or being phased out. Migrate to newer versions:

Check Google’s deprecations page for the latest information.

Pricing

Google offers generous free tiers. Check Google AI pricing for current rates.

Troubleshooting

“Quota exceeded”

• You’ve hit the free tier limit
• Wait for the quota to reset
• Or upgrade to paid tier

Image generation failed

• Ensure you’re using gemini-2.5-flash-image
• Check your prompt doesn’t violate safety guidelines
• Try a simpler prompt

Google Search not working

• Ensure the Google Search toggle is on
• Image generation models don’t support search
• Some queries may not trigger search
• Check your API key has search access

Models not loading

• Verify your API key is correct
• Check network connection
• Try refreshing the model list

“Model not found” error

• The model may have been deprecated
• Check the deprecations page
• Switch to a newer model version

Related

• Image Generation — Full guide
• OpenAI — GPT models
• xAI — Alternative image generation
• Google AI Studio — Get API keys

Sources:

• Gemini API Changelog
• Gemini API Pricing
• Gemini Models