Daniel Nkencho This workflow corresponds to n8n.io template #9247 — we link there as the canonical source.
This workflow follows the Agent → Chat Trigger recipe pattern — see all workflows that pair these two integrations.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"meta": {
"templateCredsSetupCompleted": true
},
"nodes": [
{
"id": "77530f0b-8f16-4427-a45c-54c82c66588a",
"name": "Model Selector",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1184,
1904
],
"parameters": {
"text": "={{ $json.chatInput }}",
"options": {
"systemMessage": "=# Overview\nYou are an AI agent responsible for selecting the most suitable large language model to handle a given user request. Determine whether the question is simple or complex.\n\n## Instructions\nAnalyze the user's request and return only a single number based on the complexity of the task.\n\n## Classification Rules\n- **Output 1**: Complex questions requiring deep reasoning, professional writing, research, coding, or multi-step problem solving\n- **Output 2**: Simple questions that are straightforward, conversational, or can be answered quickly\n\n## Output Format\nReturn only one of the following:\n- 1\n- 2\n\nNo explanations, no newline or spaces, no formatting, no extra text. Only the number."
},
"promptType": "define"
},
"typeVersion": 1.9
},
{
"id": "9fe14e1d-1f15-4095-b338-c96ef00a9b47",
"name": "When chat message received",
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"position": [
976,
1904
],
"parameters": {
"options": {}
},
"typeVersion": 1.3
},
{
"id": "530ca74e-ff27-4400-8527-57e9d62e3a61",
"name": "Model selector",
"type": "@n8n/n8n-nodes-langchain.modelSelector",
"position": [
1440,
2096
],
"parameters": {
"rules": {
"rule": [
{
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "f7148b18-d2be-4618-a12c-6f144f1eab14",
"operator": {
"type": "number",
"operation": "equals"
},
"leftValue": "={{ $('Model Selector').item.json.output.toNumber() }}",
"rightValue": 1
}
]
}
},
{
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "c30c1080-35d3-4cf0-a871-8034049d9d10",
"operator": {
"type": "number",
"operation": "equals"
},
"leftValue": "={{ $('Model Selector').item.json.output.toNumber() }}",
"rightValue": 2
}
]
}
}
]
}
},
"typeVersion": 1
},
{
"id": "ba211db9-bf65-4f71-9755-a47ebfea1b86",
"name": "Main Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1520,
1904
],
"parameters": {
"text": "={{ $('When chat message received').item.json.chatInput }}",
"options": {
"systemMessage": "=Here is the current date/time: {{ $now }}"
},
"promptType": "define"
},
"typeVersion": 2.2
},
{
"id": "6afd02b9-9a79-40fc-a26c-e73b837d3eda",
"name": "4.1 nano",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1488,
2352
],
"parameters": {
"model": {
"__rl": true,
"mode": "id",
"value": "=gpt-4.1-nano"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "83ca2ffd-aff1-482b-b87b-052e9f9cc193",
"name": "2.5 pro",
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"position": [
1440,
2240
],
"parameters": {
"options": {},
"modelName": "=models/gemini-2.5-pro"
},
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "9f603eb2-6ef2-43ef-8519-759eca7e91df",
"name": "2.0 flash",
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"position": [
1184,
2112
],
"parameters": {
"options": {},
"modelName": "models/gemini-2.0-flash"
},
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "c60c1f49-3732-4307-9538-86fd8b93a58c",
"name": "Overview Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
304,
1776
],
"parameters": {
"color": 4,
"width": 600,
"height": 1000,
"content": "# Adaptive LLM Router for Optimized AI Chat Responses\n\n## \ud83d\udccb What This Template Does\nThis workflow creates an intelligent chat agent that classifies incoming queries by complexity and routes them to the appropriate LLM: Gemini 2.5 Pro for complex tasks, OpenAI GPT-4.1 Nano for simple ones\u2014optimizing for cost and performance.\n\n## \ud83d\udd27 Prerequisites\n- n8n instance with LangChain nodes enabled\n- Access to Google Gemini and OpenAI APIs\n\n## \ud83d\udd11 Required Credentials\n\n### OpenAI API Setup\n1. Go to platform.openai.com/api-keys\n2. Create new secret key\n3. Add to n8n as OpenAI API credential\n\n### Google Gemini API Setup\n1. Visit makersuite.google.com/app/apikey\n2. Generate API key\n3. Add to n8n as Google PaLM API credential\n\n## \u2699\ufe0f Configuration Steps\n1. Import workflow JSON into n8n\n2. Assign credentials to LLM nodes\n3. Activate the workflow\n4. Test via chat input (use n8n's chat interface or integrate with external tools)\n\n## \ud83c\udfaf Use Cases\n- Cost-effective AI assistants for customer support\n- Dynamic query handling in educational bots\n- Optimized response generation for content tools\n- Efficient processing in research assistants\n\n## \u26a0\ufe0f Troubleshooting\n- Classification errors: Refine prompt in Model Selector\n- API issues: Check credentials and quotas\n- No response: Ensure webhook is active; test nodes individually\n- Output parsing: Verify selector outputs only 1 or 2"
},
"typeVersion": 1
},
{
"id": "00acaca1-2394-4609-bfc0-409c1f281194",
"name": "Note: Trigger",
"type": "n8n-nodes-base.stickyNote",
"position": [
928,
2096
],
"parameters": {
"width": 252,
"height": 192,
"content": "## \ud83d\udce5 When chat message received\n\n**Purpose:** For real-time chat."
},
"typeVersion": 1
},
{
"id": "fb5bad8f-6436-4a67-b0d5-fe7c2a3a58df",
"name": "Note: Classifier",
"type": "n8n-nodes-base.stickyNote",
"position": [
1152,
1712
],
"parameters": {
"color": 2,
"width": 332,
"content": "## \ud83e\udd16Model Selector\n\n**Purpose:** Classifies query complexity.\n\n**Note:** Outputs 1 (complex) or 2 (simple)."
},
"typeVersion": 1
},
{
"id": "f91cfe72-69c0-4627-b508-0159497317ca",
"name": "Note: Main Agent",
"type": "n8n-nodes-base.stickyNote",
"position": [
1744,
2048
],
"parameters": {
"color": 5,
"width": 284,
"height": 192,
"content": "## \ud83d\udcac Node: Main Agent\n\n**Purpose:** Generates response with selected model.\n\n**Note:** Includes current date in prompt."
},
"typeVersion": 1
}
],
"connections": {
"2.5 pro": {
"ai_languageModel": [
[
{
"node": "Model selector",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"4.1 nano": {
"ai_languageModel": [
[
{
"node": "Model selector",
"type": "ai_languageModel",
"index": 1
}
]
]
},
"2.0 flash": {
"ai_languageModel": [
[
{
"node": "Model Selector",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Model Selector": {
"main": [
[
{
"node": "Main Agent",
"type": "main",
"index": 0
}
]
]
},
"Model selector": {
"ai_languageModel": [
[
{
"node": "Main Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"When chat message received": {
"main": [
[
{
"node": "Model Selector",
"type": "main",
"index": 0
}
]
]
}
}
}
Credentials you'll need
Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.
googlePalmApiopenAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Elevate your AI chatbots with intelligent model selection: automatically route simple queries to cost-effective LLMs and complex ones to powerful ones, balancing performance and expenses seamlessly.
Source: https://n8n.io/workflows/9247/ — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
ModelRouter. Uses chatTrigger, agent, modelSelector, httpRequest. Chat trigger; 28 nodes.
Chat Trigger, Agent, Model Selector +8
This workflow is designed to intelligently route user queries to the most suitable large language model (LLM) based on the type of request received in a chat environment. It uses structured classifica
Chat Trigger, Agent, Model Selector +7
Ask the same question to multiple LLMs and get one answer you can actually trust. Instead of hoping one model gets it right, this workflow sends your question to four models at once, compares what the
Groq Chat, Chat Trigger, Chat +4
Transform any problem into innovative solutions using AI-powered brainstorming that combines mathematical randomness with intelligent synthesis.
Agent, Redis, Chat Trigger +2
This workflow contains community nodes that are only compatible with the self-hosted version of n8n.
Google Calendar Tool, Gmail Tool, Chat Trigger +6