> ## Documentation Index
> Fetch the complete documentation index at: https://docs.glyphformac.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Google Gemini Configuration
> Set up Google Gemini models in Glyph
Connect Glyph to Google's Gemini API to use Gemini Pro, Gemini Flash, and other models.
## Prerequisites
* Google AI Studio account: [aistudio.google.com](https://aistudio.google.com)
* API key (free tier available)
## Setup
1. Visit [Google AI Studio](https://aistudio.google.com/app/apikey)
2. Click **Get API key**
3. Create a new API key or use an existing one
4. Copy the key
Google AI Studio offers a free tier with generous limits for testing.
Go to **Settings → AI** and select the **Gemini** profile.
1. Click **Set API Key** in the authentication section
2. Paste your Google API key
3. Click **Save**
The key is stored in `.glyph/app/ai_secrets.json` in your space directory.
Click the **Model** dropdown. Glyph fetches available models from Google's API.
Popular models:
* `gemini-1.5-pro` - Most capable Gemini model
* `gemini-1.5-flash` - Fast and efficient
* `gemini-pro` - Original Gemini Pro
Open the AI panel and send a test message. You should receive a response from Gemini.
## Configuration
### Provider Settings
* **Service**: `gemini`
* **Base URL**: `https://generativelanguage.googleapis.com` (default)
* **Authentication**: API key via query parameter
### API Endpoint
Glyph uses the `/v1beta/models` endpoint to list models and sends requests to the Gemini API.
The API key is passed as a query parameter: `?key=YOUR_API_KEY`
## Model Selection
Glyph fetches the latest model list from Google's API.
### Recommended Models
| Model | Use Case | Context Window |
| ------------------ | ----------------------------- | -------------- |
| `gemini-1.5-pro` | Complex reasoning, multimodal | 2M tokens |
| `gemini-1.5-flash` | Fast tasks, high throughput | 1M tokens |
| `gemini-pro` | General purpose | 32K tokens |
**Gemini 1.5 Pro** has a **2 million token context window**, the largest of any production AI model. Attach entire codebases or books to your conversations.
### Model Naming
Google's API returns model names with a `models/` prefix (e.g., `models/gemini-1.5-pro`). Glyph automatically strips this prefix when displaying and selecting models.
## Features
### Chat Mode
Conversational interaction:
* Back-and-forth dialogue with Gemini
* No file system access
* Fast responses
* Best for Q\&A and brainstorming
### Create Mode
Gemini with workspace tools:
* **read\_file** - Read files from your space
* **search\_notes** - Search note content
* **list\_dir** - List directory contents
* Tool usage tracked in timeline view
* Best for research and knowledge retrieval
### Context Attachment
Leverage Gemini's massive context window:
* Attach files or entire folders
* Mention with `@filename` syntax
* Configure character budget (up to 250K chars)
* Gemini can handle extremely large contexts
## API Usage and Billing
### Free Tier
Google AI Studio offers a generous free tier:
* **Gemini 1.5 Flash**: 15 RPM, 1M TPM, 1,500 RPD
* **Gemini 1.5 Pro**: 2 RPM, 32K TPM, 50 RPD
RPM = requests per minute, TPM = tokens per minute, RPD = requests per day
### Paid Tier (Pay-as-you-go)
| Model | Input (per 1M tokens) | Output (per 1M tokens) |
| ---------------------- | --------------------- | ---------------------- |
| Gemini 1.5 Flash | \$0.075 | \$0.30 |
| Gemini 1.5 Pro (≤128K) | \$1.25 | \$5.00 |
| Gemini 1.5 Pro (>128K) | \$2.50 | \$10.00 |
Check current pricing at [ai.google.dev/pricing](https://ai.google.dev/pricing).
## Rate Limits
Rate limits depend on your tier and model:
* Free tier: See limits above
* Paid tier: Higher limits, see Google AI documentation
If you hit rate limits, Glyph displays the error. Wait before retrying or upgrade to paid tier.
## Troubleshooting
### "API key not set for this profile"
**Solution**: Add your Google API key in Settings → AI.
### "model list failed (400)"
**Possible causes**:
* Invalid API key
* API key doesn't have permission for Gemini API
**Solution**: Create a new API key from [AI Studio](https://aistudio.google.com/app/apikey).
### "model list failed (429)"
**Solution**: You've hit Google's rate limit. Wait before retrying or check your quota.
### Model list is empty
**Solution**: Type the model ID manually:
* `gemini-1.5-pro`
* `gemini-1.5-flash`
* `gemini-pro`
**Do not** include the `models/` prefix.
### "The model: `models/gemini-1.5-pro` does not exist"
**Cause**: You included the `models/` prefix in the model field.
**Solution**: Use just `gemini-1.5-pro` without the prefix. Glyph handles the prefix internally.
### Responses are slow with large context
**Cause**: While Gemini supports huge contexts (2M tokens), processing them takes time.
**Solution**:
* Use `gemini-1.5-flash` for faster responses
* Reduce context size if not all content is necessary
* Be patient; processing 100K+ tokens may take 10-30 seconds
## Multimodal Support
Gemini models support text, image, audio, and video inputs. However, Glyph currently only supports text inputs and outputs.
Image and multimodal support may be added in a future release.
## Security Best Practices
* Never commit `.glyph/app/ai_secrets.json` to version control
* Rotate API keys if exposed
* Monitor usage in [Google Cloud Console](https://console.cloud.google.com)
* Set up billing alerts if using paid tier
## Gemini-Specific Tips
### Large Context Use Cases
Gemini's 2M token context enables unique workflows:
* Attach entire project directories
* Include multiple books or research papers
* Provide comprehensive context for analysis
### System Instructions
Gemini respects system prompts. In create mode, Glyph adds tool usage guidelines to reduce unnecessary searches.
### Thinking Models
Google may release reasoning models (similar to OpenAI's o1). When available, they'll appear in Glyph's model list automatically.
## Next Steps
Learn about chat vs create modes
Attach large contexts with Gemini's 2M token window
Compare with Claude models
Manage multiple AI profiles