Gemini (Google)
Configure MyDeskBot to use Google's Gemini models.
Overview
Gemini is Google's family of AI models with strong multimodal capabilities.
Supported Models
- Gemini 1.5 Pro - Most capable, multimodal
- Gemini 1.5 Flash - Fast, cost-effective
- Gemini Pro - Previous generation, still capable
Get API Key
- Visit ai.google.dev
- Sign up or log in with Google account
- Go to API Keys section
- Create a new API key
- Copy the key
Configuration Options
Model Selection
| Model | Best For | Cost | Speed |
|---|---|---|---|
| Gemini 1.5 Pro | Complex tasks, multimodal | Medium | Fast |
| Gemini 1.5 Flash | Fast responses | Low | Very Fast |
| Gemini Pro | General tasks | Low | Medium |
Parameters
- Temperature (0-2) - Controls randomness
- Max Tokens - Maximum response length
- Top K - Limits token choices
- Top P - Nucleus sampling
Pricing
Check Google AI Pricing for current rates.
Rate Limits
Gemini has rate limits based on your plan:
- Free Tier - Limited usage per day
- Pay-as-you-go - Higher limits
- Enterprise - Custom limits
Monitor usage at ai.google.dev.
Why Choose Gemini?
- Multimodal - Excellent at handling images and text
- Large Context - Up to 1M tokens context window (1.5 Pro)
- Google Ecosystem - Integrates well with Google services
- Competitive Pricing - Good value for cost
Troubleshooting
Invalid API Key
- Verify your API key is correct
- Check Google Cloud project status
- Enable the Gemini API in Google Cloud Console
Quota Exceeded
- Check your usage quota
- Request quota increase if needed
- Implement retry logic