🧠 Models vs Agents

Understanding their roles in n8n trading automation

Learning Objectives:

🧠 Understand Models (LLM Chain) vs Agents (AI Agent Node)
💬 Learn how chat triggers work with Models
📊 Explore interactive workflow infographics
⚡ Know when to use each approach in n8n

🎯 The Context in n8n

In n8n trading automation, we have two distinct approaches:

🧠 Models (LLM Chain)
Reactive analysis systems triggered by requests or chat

🤖 Agents (AI Agent Node)
Autonomous systems that use tools and make decisions

n8n Implementation:
• Models: Use LLM Chain nodes, triggered by webhooks, chat, or manual triggers
• Agents: Built using AI Agent nodes with tool integrations

🧠 Models in n8n (LLM Chain)

Definition: Reactive systems using LLM Chain nodes that respond to requests and provide structured analysis.

💬 Chat Triggered: Can be triggered by chat interfaces in n8n
🔗 LLM Chain: Uses LLM Chain nodes for processing
📊 Process: Request → Fetch data → LLM analysis → Standardized output
🔄 Integration: Outputs feed other workflows or respond to chat

Key Characteristics:
• Waits for external requests to start working
• Provides analysis and recommendations
• Returns structured data for other systems to use
• Perfect for on-demand market analysis

📤 Model Output Example

Here's what a Model typically outputs when analyzing GBPUSD:

{ "symbol": "GBPUSD", "analysis": { "trend": "bullish", "confidence": 0.78, "signals": { "rsi": "oversold_recovery", "ma_cross": "golden_cross" } }, "recommendation": "BUY", "entry_price": 1.3485, "stop_loss": 1.3450, "take_profit": 1.3520, "risk_reward_ratio": 2.1, "timestamp": "2025-01-08T10:30:00Z", "chat_response": "📈 GBPUSD showing strong bullish signals with golden cross confirmation" }

                This structured output can be:

                • Sent back to a chat interface

                • Used by another n8n workflow

                • Stored in a database

                • Sent to external systems via API

🤖 Agents in n8n (AI Agent Node)

Definition: Autonomous systems built with AI Agent nodes that use tools, make decisions, and take actions independently.

🤖 AI Agent Node: Core node that provides autonomous behavior
🛠️ Tool Integration: Can use multiple tools (MT5, calculators, APIs)
⏰ Scheduled: Runs on intervals or event triggers
🧠 Memory: Maintains context and learns from actions

Key Characteristics:
• Runs continuously on a schedule
• Makes autonomous trading decisions
• Uses multiple tools to gather information and execute trades
• Learns from experience and adapts behavior

📊 Interactive Workflow Infographics

Click the buttons to explore the different workflow structures:

💬
Chat/Request
Trigger

→

🔍
Fetch Input
Data

→

🧠
LLM Chain
Analyze

→

📋
Standardize
Output

→

📤
Response
Output

🤖
AI Agent
Core

⏰ Schedule Trigger

📊 Fetch Analysis

📈 Check Status

🧠 Understand Environment

🎯 Determine Actions

🛠️ Use Tools

📝 Store Memory

🎯 Key Takeaways for n8n

🧠 Use Models When:
• Building chat-based trading advisors
• Need on-demand market analysis
• Want structured responses for other systems
• Creating analysis APIs
• Human oversight required

🤖 Use Agents When:
• Want fully automated trading
• Need continuous monitoring
• Require tool orchestration
• Want adaptive behavior
• Building autonomous systems

🔧 n8n Implementation Guide:
• Models: Start with Chat Trigger → LLM Chain → Respond
• Agents: Use Schedule Trigger → AI Agent → Tool integrations
• Hybrid: Agent can call Model workflows as analysis tools

⚡ Decision Framework:
• Interactive analysis? → Model with chat triggers
• Autonomous trading? → Agent with AI Agent node
• Both? → Agent that uses Model as analysis tool

Start with Models for learning, evolve to Agents for automation!