Skip to content

Configuration

This page provides information about configuring VT.ai to suit your needs.

Configuration Options

VT.ai offers several configuration options that can be set through various methods:

  1. Command-line arguments
  2. Environment variables
  3. UI settings
  4. Configuration files

Model Configuration

You can configure which models to use for different types of operations:

  • Chat models
  • Vision models
  • Text-to-speech models
  • Image generation models

Web Search Settings

Customize how VT.ai handles web searches:

  • Summarize Search Results: Toggle between AI-summarized results and raw search output
  • Enabled (default): Synthesizes multiple search results into a coherent summary
  • Disabled: Shows raw search results with minimal processing

  • Include URLs: Choose whether to include source URLs in the search results

  • Useful for citation and verification

  • Search Context Size: Control the depth of web searches

  • Options: low, medium, high
  • Higher values search more deeply but may take longer

Performance Settings

Fine-tune the performance of your models:

  • Temperature
  • Top-P
  • Max tokens
  • Response format

Routing Configuration

Configure the semantic routing system:

  • Enable/disable dynamic routing
  • Customize routing thresholds
  • Add custom intents

API Key Management

Manage your API keys for different providers:

  • OpenAI
  • Anthropic
  • Google
  • Other providers

UI Customization

Customize the user interface to your preferences:

  • Theme settings
  • Display options
  • Keyboard shortcuts

This page is under construction. More detailed configuration information will be added soon.