Docs
AI Chat Configuration

AI Chat Configuration

Learn how to set up and configure AI chat after your first login, including enabling smart AI chat features for an enhanced experience.

Getting started with AI chat is simple and straightforward. This guide will walk you through the complete setup process, from configuring your first AI provider to enabling advanced smart chat features.

Quick Start

After logging in for the first time, follow these steps to get AI chat up and running:

  1. Navigate to AI Settings → Go to Settings > AI
  2. Add API Keys → Configure at least one AI provider
  3. Test Your Setup → Start a basic chat conversation
  4. Optimize (Optional) → Enable Smart Chat for enhanced features

Step 1: Configure AI Provider Keys

Before you can use AI chat, you need to configure at least one AI provider by adding their API key.

Supported Providers

We support several leading AI providers:

ProviderModels AvailableKey FormatGet API Key
OpenAIGPT-4, GPT-3.5, GPT-4osk-...OpenAI Platform
AnthropicClaude 3, Claude 3.5sk-ant-...Anthropic Console
Google AIGemini Pro, Gemini FlashAIza...Google AI Studio
Azure OpenAIAzure-hosted modelsCustom formatAzure Portal
OllamaLocal open-source modelsOptional tokenOllama Docs

Adding Your First API Key

  1. Go to Settings → AI → API Keys
  2. Select a Provider from the dropdown
  3. Enter Your API Key (obtained from the provider's website)
  4. Add a Name (e.g., "My OpenAI Key")
  5. Click "Add Key"
# Example: Getting OpenAI API Key
1. Visit https://platform.openai.com/api-keys
2. Click "Create new secret key"
3. Copy the key (starts with sk-...)
4. Paste it into the platform

Key Management Tips

  • Keep Keys Secure: API keys are encrypted and stored safely
  • Test Keys: The system automatically validates keys when added
  • Multiple Providers: You can configure multiple providers for different models
  • Usage Tracking: Monitor your API usage through provider dashboards

Step 2: Configure AI Settings

Once you have at least one provider configured, customize your AI chat experience:

Model Selection

  • Default Model: Choose your preferred AI model
  • Auto-Selection: The system suggests the best available model
  • Model Switching: Change models anytime during conversations

Conversation Settings

  • Temperature: Control creativity (0.1 = focused, 0.9 = creative)
  • Max Tokens: Set response length limits
  • System Prompt: Define AI personality and behavior
  • Web Search: Enable real-time information access

Example Configuration

Default Model: gpt-4o
Temperature: 0.7
Max Tokens: 4000
System Prompt: 'You are a helpful AI assistant with expertise in technology and business.'
Web Search: Enabled

Step 3: Your First Chat

With providers configured, you're ready to chat:

  1. Navigate to AI → Chat
  2. Start typing your message
  3. Press Enter to send
  4. Wait for response from your selected AI model

Chat Features

  • Image Support: Upload images with vision-capable models
  • Voice Input: Use speech-to-text (Ctrl+Shift+V)
  • Conversation Memory: Chat history is automatically saved
  • Model Switching: Change models mid-conversation
  • Web Search Toggle: Enable/disable per conversation

Smart Chat automatically optimizes your AI experience with advanced features. It's perfect for users who want the best possible AI interaction without manual configuration.

What Smart Chat Includes

Intelligent Model Selection - Automatically chooses the best available model
Enhanced Knowledge Retrieval (RAG) - Access to your document knowledge base
Real-time Web Search - Current information and fact-checking
Web Content Scraping - Extract information from web pages
Advanced Reasoning (MCP) - Enhanced problem-solving capabilities
Optimized Settings - Perfect temperature, token limits, and prompts

Enabling Smart Chat

Option 1: Smart Chat Button

  1. Look for the "Optimize" button with sparkles icon ✨
  2. Click "Optimize & Start" when prompted
  3. Let the system configure optimal settings
  4. Start chatting with enhanced capabilities

Option 2: Manual Configuration

If you prefer manual setup:

  1. Navigate to Settings → AI → Chat Settings
  2. Select a high-quality model (GPT-4, Claude, or Gemini)
  3. Enable Web Search for current information
  4. Set up RAG (Settings → AI → Knowledge Base)
  5. Enable MCP for advanced reasoning

Smart Chat Benefits

FeatureStandard ChatSmart Chat
Response QualityGoodExcellent
Knowledge AccessModel training onlyLive web + documents
Problem SolvingBasicAdvanced reasoning
Setup TimeManual configurationOne-click optimization
Current InformationLimited to training dataReal-time web access

Advanced Features

Web Search Integration

Enable real-time information access:

  • Toggle per conversation with the Web Search badge
  • Automatic fact checking for current events
  • Source citations when information is found online

Knowledge Base (RAG)

Upload documents for AI to reference:

  1. Go to Settings → AI → Knowledge Base
  2. Upload documents (PDF, Word, text files)
  3. Enable RAG in chat settings
  4. AI will reference your documents in responses

Custom System Prompts

Personalize AI behavior:

You are an expert software architect specializing in scalable systems.
Provide detailed technical guidance with code examples when appropriate.
Always consider security, performance, and maintainability.

Multi-Model Conversations

  • Switch models mid-conversation
  • Compare responses from different AI providers
  • Leverage strengths of different models

Troubleshooting

Common Issues

"Model not available"

  • Check if API key is correctly configured
  • Verify the key has credits/quota remaining
  • Try a different model from the same provider

"API key configuration required"

  • Go to Settings → AI → API Keys
  • Add at least one valid provider key
  • Ensure the key is active and not expired

"Chat not responding"

  • Check your internet connection
  • Verify API provider status
  • Try refreshing the page

Getting Help

  • Check provider status pages for outages
  • Monitor usage quotas on provider dashboards
  • Review error messages for specific guidance
  • Try different models if one isn't working

Best Practices

Security

  • Never share API keys with others
  • Use separate keys for different projects
  • Monitor usage regularly
  • Rotate keys periodically

Cost Management

  • Set monthly quotas in provider settings
  • Monitor token usage for expensive models
  • Use appropriate models for different tasks
  • Enable Smart Chat for optimized efficiency

Performance

  • Choose models wisely based on task complexity
  • Adjust temperature for desired creativity level
  • Set reasonable token limits to control costs
  • Use web search selectively for current info needs

Next Steps

Once you have AI chat configured:

  1. Explore different models to find your favorites
  2. Upload documents to enhance knowledge base
  3. Experiment with system prompts for specialized tasks
  4. Join the community to share tips and tricks

Ready to start your AI conversation? Head to the Chat section and begin exploring the possibilities!