Step 4 - Model Config

Step 4: Model Configuration ⚙️

Fine-tuning your AI's performance

Understanding AI Models:

Think of this like choosing the "brain power" for your chatbot. More powerful models give better answers but cost more per conversation.

🧠 AI Model Selection

Select from the available models given by your AI provider:

Setting the Parameters :

💬 Max Messages (Default: 5)

• What it controls: How much chat history the AI remembers

• Higher numbers: Better context understanding, higher costs

• Lower numbers: Less context, lower costs

• Sweet spot: 3-7 messages for most use cases

📄 Max Context Chunks (Default: 3)

• What it controls: How many knowledge base documents the AI can reference per response

• Higher numbers: More comprehensive answers, higher costs

• Lower numbers: Focused answers, lower costs

• Sweet spot: 2-5 chunks for most use cases

🔤 Max Tokens (Default: 1000)

• What it controls: Maximum length of AI responses

• 1000 tokens ≈ 750 words

• Recommendations:

  • Short answers: 500 tokens

  • Detailed responses: 1000-1500 tokens

  • Very detailed: 2000+ tokens

💰 Cost Optimization Tips:

1. Start conservative: Use default settings initially

2. Monitor usage: Check your AI provider's usage dashboard

3. Adjust based on needs: Increase limits only if responses are too short/limited

4. Test thoroughly: Higher settings = higher costs per conversation

✅ Step 4 Checklist:

• Select appropriate AI model for your use case

• Set conversation memory (3-7 messages recommended)

• Configure context chunks (2-5 recommended)

• Set response length (500-1500 tokens recommended)