​
Node Description

The LLM node allows users to input specific instructions, which are processed by a language model to generate text, provide analysis, or execute defined tasks.

​
Key Features

  • Customizable Instructions: Tailor the model’s response with detailed instructions and dynamic variables (e.g., {{myVariable}}).
  • Adjustable AI Parameters: Fine-tune model behavior by configuring temperature, token limits, and more.
  • Versatile Applications: Use for text generation, analysis, summarization, and other NLP tasks.

​
Node Inputs

​
Required Fields

Instructions
Provide specific instructions to the model. Use dynamic variables ({{myVariable}}) to personalize input.
Examples:
  • "Write a story about F1 driver loosing the championship for this driver {{driver}} with the team {{team}}"
  • "Summarize the provided text in 100 words."

​
Optional Fields

System Message (Optional)
Set the context for the AI model to operate under.
Example: "You are a data analyst specializing in market trends."

​
AI Settings

​
Model

Choose the LLM to use.
Example: "OPENAI - gpt-4o-2024-05-13"

​
Temperature

Controls the randomness of the output:
  • 0.2: Deterministic and focused.
  • 0.6: Balanced between creativity and precision (default).
  • 0.8: More creative and exploratory.

​
Max Output Tokens

Defines the length of the response.
Example: 1000 tokens.

​
Top K

Limits the selection to the top K most likely tokens.
Example: 40

​
Top P

Adjusts probability for token selection.
Default: 1 (uses full probability distribution).

​
Node Output

​
Output

The LLM’s response based on the provided instructions.

​
Example Output:

{
  "summary": "AI tools streamline workflows and enhance productivity in various industries.",
  "keywords": ["AI tools", "productivity", "workflows"]
}

​
Reflection Agent Integration

​
Purpose

The Reflection Agent works as a guide to refine AI-generated content by providing constructive feedback.

​
Best Practices

  • Start with broad guidance and refine with focused instructions.
  • Ensure instructions are clear and do not contradict.
  • Use feedback iteratively for improved content quality.

​
Example Usage

  1. Content Analysis
    Instruction: "Analyze if the text aligns with the user's search intent."
  2. SEO Suggestions
    Instruction: "Recommend additional keywords for optimizing this content for search engines."
  3. Structural Review
    Instruction: "Check if the content is well-structured with clear headings and subheadings."
The LLM node is adaptable for a wide range of use cases, making it an essential tool for generating, refining, and analyzing content.