Setup Guide

Quick Setup Process

Once you have your API key from your chosen provider, follow these simple steps to connect it to Nexalexica:

Step 1: Navigate to Settings

1. Open Nexalexica
2. Go to Settings (usually in the top menu or sidebar)
3. Scroll down to find the “API Keys” section

Step 2: Add Your API Key

1. Locate your chosen provider in the API Keys section
2. Copy your API key from the provider’s dashboard
3. Paste it into the corresponding input field for your provider

Step 3: Select Models (Automatic)

After adding your API key:

• Chat models will automatically load from your provider
• Embedding models will also fetch automatically through our endpoint
• Choose your preferred models from the dropdown menus

Important Model Selection Guidelines

Current Limitations

Nexalexica currently works best with smaller models due to context window limitations. We recommend:

✅ Recommended Models by Provider

Google Gemini:

• Choose Gemini Pro 1.0 or Gemini 1.5 instead of Gemini 2.0
• These provide optimal functionality with current system

OpenAI:

• Use GPT-4 or GPT-4 Turbo for best performance
• GPT-3.5 Turbo works well for faster responses

Anthropic:

• Claude 3 Haiku or Claude 3 Sonnet are ideal choices
• Avoid Claude 3 Opus for now due to size limitations

Other Providers:

• Select models with smaller context windows (8K-32K tokens)
• Choose 7B-13B parameter models for optimal performance

Why Smaller Models?

• Faster response times
• Better compatibility with all Nexalexica features
• More stable performance across different search modes
• Optimal resource usage

Troubleshooting

Models Not Loading?

• Double-check your API key is correct
• Ensure you have sufficient credits/quota with your provider
• Try refreshing the page after adding the key

Performance Issues?

• Switch to a smaller, recommended model
• Check your internet connection
• Verify your provider’s service status

Next Steps

Once your API is configured:

1. Test with a simple search query
2. Try different search modes (Normal, Academic, etc.)
3. Experiment with various model combinations
4. Adjust settings based on your needs

Ready to start searching smarter? Your API setup is complete!