3 Game-Changing NEW AI Agent Updates in n8n 🔥

Key Concepts:

• Fallback Chat Model
• AI Agent as a Tool
• Model Selector
• Orchestration Layer
• Lang Chain Node
• VPS (Virtual Private Server)
• LLMs (Large Language Models)

1. Fallback Chat Model

• Main Point: N9's new update introduces a fallback chat model feature for AI agents, providing redundancy in case the primary model fails.
• Specific Details:
  • The "enable fallback model" option is found within the AI agent settings.
  • Users can specify a primary chat model (e.g., OpenAI GPT-4) and a fallback model (e.g., Claude).
  • If the primary model fails (e.g., due to an incorrect API key), the AI agent automatically switches to the fallback model.
• Example: The presenter demonstrates this by intentionally using an incorrect API key for the primary OpenAI model, causing the AI agent to switch to the Claude fallback model.
• Real-world Application: Useful for production workflows where uninterrupted service is critical, such as client automations. It prevents workflow failures due to API key issues or other problems with the primary model.
• Benefit: Ensures workflow automation continues even if the primary AI model fails, allowing time to fix the issue without disrupting the entire process.

2. AI Agent as a Tool

• Main Point: AI agents can now be attached to other AI agents as tools, creating an orchestration layer for more complex workflows.
• Specific Details:
  • Within an AI agent, you can add another AI agent as a tool.
  • This allows the primary AI agent to delegate specific tasks to other specialized AI agents.
  • Each AI agent tool can have its own memory, tools, and chat model.
• Example: The presenter refers to a previous video demonstrating an "AI agent army" where an orchestration agent uses tools like an email agent, calendar agent, calculator, and company knowledge agent.
• Real-world Application: Building complex AI systems where different agents handle specific tasks, such as scheduling meetings (calendar agent) or sending emails (email agent).
• Benefit: Simplifies complex workflows, reduces hallucination (inaccurate or nonsensical outputs), and makes it easier to manage and maintain AI systems.

3. Model Selector

• Main Point: The model selector allows users to add up to 10 different models and dynamically switch between them based on specific conditions.
• Specific Details:
  • The model selector node allows attaching multiple models (up to 10).
  • This is useful for using smaller, more efficient models for specific tasks and larger models only when necessary.
• Example: The presenter references a workflow created by Mario that uses a "switch model" node (similar to the model selector) to handle customer complaints. The workflow uses a less capable model initially and switches to a more capable model if the response is not satisfactory.
• Alternative Implementation: The presenter also mentions using a Lang Chain node with custom code for more control over model selection.
• Hosting N9 on a VPS:
  • The presenter recommends Hostinger VPS for self-hosting N9, especially for using Lang Chain nodes.
  • He provides a step-by-step guide on setting up N9 on Hostinger VPS, including selecting a plan (KVM2 recommended), applying a coupon code ("AIWORKSHOP" for 10% off), and installing N9.
• Real-world Application: Optimizing AI workflows by using the most appropriate model for each task, reducing costs and improving performance.
• Benefit: Provides flexibility in choosing the right model for the job, allowing for cost optimization and improved accuracy.

4. Monetizing AI Agents

• Main Point: The presenter promotes a community and course on how to monetize AI agents, including starting an AI agency, acquiring clients, and pricing services.
• Specific Details:
  • The community offers resources on starting an AI agency with N9, including how to get the first client and run discovery calls.
  • It also provides examples of proposals and pricing strategies.
  • A voice AI course with certification is also available.
• Benefit: Provides practical guidance and resources for individuals looking to build a business around AI automation.

5. Notable Quotes:

• "This is a super powerful update, especially if you're creating uh workflows that might require additional uh kind of backup plans, especially for production." (Regarding the fallback chat model)
• "With this new update, it makes things very easy and also reduces a lot of the hallucination." (Regarding AI Agent as a Tool)
• "You basically identify inside your uh model selector when to use these different models." (Regarding the Model Selector)

6. Technical Terms:

• Fallback Chat Model: A backup AI model that is used when the primary model fails.
• AI Agent as a Tool: The ability to use one AI agent as a tool within another AI agent, creating a hierarchical structure.
• Model Selector: A node that allows users to dynamically switch between multiple AI models based on specific conditions.
• Orchestration Layer: A system that manages and coordinates the interaction between multiple AI agents.
• Lang Chain Node: A node that allows users to integrate Lang Chain, a framework for building language model applications, into N9 workflows.
• VPS (Virtual Private Server): A virtual machine that provides dedicated resources for hosting applications.
• LLMs (Large Language Models): Powerful AI models that can generate human-like text.

7. Logical Connections:

• The fallback chat model provides redundancy, ensuring that workflows continue to function even if the primary AI model fails.
• AI Agent as a Tool allows for the creation of complex workflows by breaking down tasks into smaller, more manageable components handled by specialized AI agents.
• The model selector enables dynamic switching between different AI models, optimizing performance and cost.
• The presenter connects the use of Lang Chain nodes with the need for a VPS, as Lang Chain nodes often require a self-hosted environment.

8. Data/Statistics:

• The model selector allows attaching up to 10 models.
• The AIWORKSHOP coupon code provides an additional 10% off on Hostinger VPS plans.

9. Conclusion:

The three updates to N9's AI agents – the fallback chat model, AI agent as a tool, and model selector – significantly enhance the platform's capabilities for building robust, efficient, and cost-effective AI automation workflows. The fallback chat model ensures redundancy, AI agent as a tool enables complex orchestration, and the model selector optimizes model usage. The presenter also provides practical guidance on self-hosting N9 on a VPS and monetizing AI agent skills.