Skip to content

Configuration Guide

This guide covers the essential configuration steps for AI Complete Suite, focusing on AI provider setup and basic system configuration.

⚙️ Overview

Before you can use AI assistants, you need to configure the AI providers (ChatGPT, Claude, Gemini) that will power your AI system. This guide walks you through the configuration process step by step.

📋 Prerequisites

Before starting configuration, ensure you have:

  • Installed all required Python dependencies
  • Set up PostgreSQL with pgvector extension
  • Installed the AI Complete Suite modules
  • Obtained API keys from your chosen AI providers

🔧 Step 1: AI Provider Configuration

Understanding AI Configurations

AI configurations define how your system connects to external AI providers. Each configuration contains:

  • Connection settings for the AI provider
  • Model selection and parameters
  • Security settings for API keys
  • Performance tuning options

Creating AI Configurations

  1. Access Configuration Settings

    • Go to Settings → Technical → AI → Configurations
    • Click Create to add a new configuration

  2. Basic Configuration Settings

    • Name: Give your configuration a descriptive name (e.g., "ChatGPT Production", "Claude Support")
    • Type: Select your AI provider type
    • API Key: Enter your provider's API key (stored securely)
    • Active: Enable/disable this configuration

  3. Advanced Parameters

    • Model: AI model to use (auto-filled based on type)
    • Temperature: Controls response randomness (0.0-2.0)
      • Lower values (0.1-0.3): More focused, consistent responses
      • Higher values (0.7-1.0): More creative, varied responses
    • Max Tokens: Maximum response length (default: 2000)
    • Allow Upload Files: Enable file attachments (varies by provider)

Provider-Specific Configuration

OdooAI (Default Provider)

  • Required Settings:

    • Type: OdooAI
    • API Key: Not required (free provider)
    • Model: Auto-configured
    • File Support: Not available

ChatGPT (OpenAI) Configuration

Claude (Anthropic) Configuration

Gemini (Google) Configuration

Azure OpenAI Configuration

  • Required Settings:

    • Type: Azure OpenAI
    • API Key: From your Azure OpenAI resource
    • Azure Endpoint: The base URL of your Azure OpenAI resource (e.g., https://your-resource-name.openai.azure.com/)
    • Azure Deployment: The name of your model deployment in Azure (as configured in the Azure Portal)
    • Azure API Version: The API version to use (e.g., 2024-02-15-preview)

  • Recommended Settings:

    • Temperature: 0.3-0.7 (similar to ChatGPT)
    • Max Tokens: 2000-4000

  • Vision Support:

    If your Azure deployment runs a vision-enabled model (like GPT-4 Turbo with Vision), you can send images and PDF attachments in conversations.

Provider Comparison

Feature OdooAI ChatGPT Claude Gemini Azure OpenAI
Cost Free Pay-per-use Pay-per-use Pay-per-use Pay-per-use
API Key Not required Required Required Required Required
File Support None Images only Images + PDFs Images only Images + PDFs (if vision-enabled)
Best For Testing, basic use General purpose Complex reasoning Creative tasks Enterprise deployments, custom models
Response Quality Basic Excellent Excellent Very good Depends on deployed model
Speed Fast Fast Medium Very fast Varies
Context Length Limited Large Very large Large Depends on deployed model
Setup Complexity Simple Medium Medium Medium Medium-High
Web Search No Yes Yes No No

Recommendations:

  • Start with OdooAI for testing and basic functionality
  • Use ChatGPT for general business applications
  • Choose Claude for complex reasoning and document analysis
  • Pick Gemini for creative content and fast responses
  • Use Azure OpenAI if you have your own Azure subscription or need to run models on your own infrastructure

⚡ Step 2: Performance Optimization

Advanced Configuration

Temperature Settings:

  • 0.0-0.2: Highly deterministic, consistent responses
  • 0.3-0.5: Balanced, good for most use cases
  • 0.6-0.8: More creative, varied responses
  • 0.9-1.0: Very creative, less predictable

Token Limits:

  • 1000-2000: Short, concise responses
  • 2000-4000: Standard responses
  • 4000+: Long, detailed responses

🧪 Step 3: Testing Your Configuration

Basic Connection Test

  1. Verify Configuration

    • Check that configuration is active
    • Verify API key is valid
    • Test basic connectivity

  2. Test with Simple Query

    • Create a basic AI assistant
    • Send a simple test message
    • Verify response quality

🚀 Next Steps

After completing configuration:

  1. Create AI Contexts - Define AI behavior and personality
  2. Set up Data Sources - Configure knowledge bases
  3. Create AI Assistants - Build virtual agents
  4. Test Integration - Verify Discuss and Livechat integration

🆘 Support

If you encounter configuration issues:

  • Check the troubleshooting guide for detailed solutions
  • Review system logs for error messages
  • Contact support with specific error details
  • Verify all prerequisites are met

This configuration guide provides the foundation for setting up AI providers. Once configured, you can proceed to create contexts, data sources, and AI assistants.