ChatbotBuilder logo

4 min read

What This Does #

Lets you select which AI model powers your bot, affecting response quality, speed, and capabilities.

When to Use This #

  • Setting up a new bot
  • Optimizing for specific needs (speed vs. quality)
  • Improving bot performance
  • Adjusting for high-volume usage

Step-by-Step Instructions #

  1. Access AI Settings
    • Go to Build → AI Settings
    • You’ll see the AI Model selection dropdown
AI Model from AI Setting
  1. Review Available Models
    • See the list of available AI models
    • Each has different strengths and speeds
  2. Select Your Model
    • Click the dropdown to see options
    • Choose based on your specific needs
AI Model Implementation
  1. Test Performance
    • Save your selection
    • Test the bot to see how it performs
    • Adjust if needed

Available Models Explained #

Google Gemini 2.0 Flash

  • Best for: Most general business uses
  • Speed: Very fast responses
  • Quality: High-quality, natural conversations
  • Good for: Customer service, sales support, general inquiries

Google Gemini 2.0 Flash Lite

  • Best for: High-volume, simple interactions
  • Speed: Extremely fast
  • Quality: Good for straightforward questions
  • Good for: Basic FAQ bots, simple information lookup

OpenAI GPT-4.1

  • Best for: Complex conversations requiring deep understanding
  • Speed: Slower but very thorough
  • Quality: Highest quality responses
  • Good for: Complex customer service, detailed explanations

OpenAI GPT-4.1 Mini

  • Best for: Balanced performance and cost
  • Speed: Moderate speed
  • Quality: Good quality responses
  • Good for: Standard business applications

Anthropic 4 Sonnet (Recommended)

  • Best for: Balanced performance for a wide range of complex tasks
  • Speed: Slower but very thorough
  • Quality: Highest quality responses
  • Good for: Creative content generation, detailed analysis, complex problem-solving

Anthropic 3.7 Sonnet

  • Best for: General-purpose use with a focus on speed and efficiency for everyday tasks
  • Speed: Slower but very thorough
  • Quality: High, but potentially less nuanced than Anthropic 4
  • Good for: Summarization, drafting emails, moderate complexity coding assistance

Anthropic 3.5 Sonnet

  • Best for: Efficient and reliable performance for common language tasks
  • Speed: Fast
  • Quality: Good quality responses
  • Good for: Customer service interactions, simple content creation, data extraction

Anthropic 3.5 Haiku

  • Best for: High-volume, simple interactions
  • Speed: Extremely fast
  • Quality: Good for straightforward questions
  • Good for: Basic FAQ bots, simple information lookup, quick responses

Grok 3

  • Best for: Advanced reasoning and comprehensive knowledge
  • Speed: Fast
  • Quality: Very High
  • Good for: Deep dives, complex problem-solving, nuanced understanding

Grok 3 Fast

  • Best for: Prioritizing speed for general use
  • Speed: Extremely Fast
  • Quality: High, with some potential trade-offs for speed
  • Good for: Real-time applications, quick information retrieval

Grok 3 Mini

  • Best for: Lightweight, efficient performance for mobile or resource-constrained environments
  • Speed: Very Fast
  • Quality: Good for common tasks
  • Good for: Mobile applications, simple chatbot interactions, quick drafting

Grok 3 Mini Fast

  • Best for: Maximum speed for very simple, high-volume requests in constrained environments
  • Speed: Utmost Fast
  • Quality: Good for straightforward queries, potentially less depth
  • Good for: High-throughput simple tasks, quick checks

How to Choose the Right Model #

Customer Service

  • High volume, simple queries: Gemini 2.0 Flash Lite, Anthropic 3.5 Haiku
  • Standard volume: Gemini 2.0 Flash, Anthropic 3.5 Sonnet
  • Complex issues: GPT-4.1, Anthropic 4 Sonnet, Claude 3.7 Sonnet
  • Emotional/nuanced support: Anthropic 4 Sonnet, GPT-4.1

Sales & Marketing

  • Product information: Gemini 2.0 Flash, Anthropic 3.5 Sonnet
  • Consultative selling: GPT-4.1, Anthropic 4 Sonnet, Claude 3.7 Sonnet
  • Lead qualification: Gemini 2.0 Flash, GPT-4.1 Mini
  • Creative content: Anthropic 4 Sonnet, GPT-4.1

Technical Support

  • Simple troubleshooting: Gemini 2.0 Flash, Anthropic 3.5 Sonnet
  • Complex problem-solving: GPT-4.1, Claude 3.7 Sonnet, Grok 3
  • Code assistance: GPT-4.1, Claude 3.7 Sonnet, Anthropic 4 Sonnet
  • Documentation queries: GPT-4.1 Mini, Anthropic 3.5 Sonnet

Development & Coding

  • Code generation: GPT-4.1, Claude 3.7 Sonnet, Anthropic 4 Sonnet
  • Code review: Claude 3.7 Sonnet, GPT-4.1
  • Simple scripts: GPT-4.1 Mini, Gemini 2.0 Flash
  • Complex debugging: Claude 3.7 Sonnet, Grok 3

Research & Analysis

  • Quick research: Gemini 2.0 Flash, Grok 3
  • Deep analysis: Anthropic 4 Sonnet, GPT-4.1, Grok 3
  • Real-time information: Grok 3 (with X integration)
  • Long documents: Claude 3.7 Sonnet, GPT-4.1

Performance vs. Cost Framework #

Speed Tier (Fastest Response)

  • Best for: High-volume interactions, simple queries, real-time chat
  • Models: Gemini 2.0 Flash Lite, Anthropic 3.5 Haiku, Grok 3 Mini
  • Cost: Lowest

Balanced Tier (Speed + Quality)

  • Best for: Standard business applications, moderate complexity
  • Models: Gemini 2.0 Flash, Anthropic 3.5 Sonnet, GPT-4.1 Mini
  • Cost: Medium

Premium Tier (Highest Quality)

  • Best for: Complex reasoning, nuanced understanding, critical applications
  • Models: GPT-4.1, Anthropic 4 Sonnet, Claude 3.7 Sonnet, Grok 3
  • Cost: Highest

Testing Your Choice #

  • Try different models with the same questions
  • Check response speed and quality
  • Get feedback from users
  • Monitor analytics for performance changes
  • Switch models if needed

Tips for Success #

  • Start with Gemini 2.0 Flash for most use cases
  • Upgrade to GPT-4.1 if you need better understanding
  • Use Lite versions for high-volume, simple bots
  • Test thoroughly before going live
  • Monitor performance and adjust as needed