This workflow follows the Chainllm → 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 →
{
"id": "dQC8kExvbCrovWf0",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Dynamically switch between LLMs Template",
"tags": [],
"nodes": [
{
"id": "962c4b29-c244-4d68-93e1-cacd41b436fc",
"name": "When chat message received",
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"position": [
220,
80
],
"parameters": {
"options": {}
},
"typeVersion": 1.1
},
{
"id": "6fc4f336-09e3-4e79-94e9-e5eff04e4089",
"name": "Switch Model",
"type": "@n8n/n8n-nodes-langchain.code",
"position": [
540,
320
],
"parameters": {
"code": {
"supplyData": {
"code": "let llms = await this.getInputConnectionData('ai_languageModel', 0);\nllms.reverse(); // reverse array, so the order matches the UI elements\n\nconst llm_index = $input.item.json.llm_index;\nif (!Number.isInteger(llm_index)) {\n console.log(\"'llm_index' is udefined or not a valid integer\");\n throw new Error(\"'llm_index' is udefined or not a valid integer\");\n}\n\nif(typeof llms[llm_index] === 'undefined') {\n console.log(`No LLM found with index ${llm_index}`);\n throw new Error(`No LLM found with index ${llm_index}`);\n}\n\nreturn llms[llm_index];"
}
},
"inputs": {
"input": [
{
"type": "ai_languageModel",
"required": true
}
]
},
"outputs": {
"output": [
{
"type": "ai_languageModel"
}
]
}
},
"typeVersion": 1
},
{
"id": "68511483-355b-45c1-915f-e7517c42b809",
"name": "Set LLM index",
"type": "n8n-nodes-base.set",
"position": [
440,
80
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "24b4d30e-484a-4cc1-a691-0653ed764296",
"name": "llm_index",
"type": "number",
"value": "={{ $json.llm_index || 0 }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "adc2f24c-0ad6-4057-bb3b-b46563c72ee8",
"name": "Increase LLM index",
"type": "n8n-nodes-base.set",
"position": [
1420,
-200
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "24b4d30e-484a-4cc1-a691-0653ed764296",
"name": "llm_index",
"type": "number",
"value": "={{ $('Set LLM index').item.json.llm_index + 1 }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "eace2dd7-9550-47ba-a4c3-4f065f80757b",
"name": "No Operation, do nothing",
"type": "n8n-nodes-base.noOp",
"position": [
1640,
540
],
"parameters": {},
"typeVersion": 1
},
{
"id": "c1735d1c-5dc4-4bd5-9dde-3bb04b8811c3",
"name": "Check for expected error",
"type": "n8n-nodes-base.if",
"position": [
1040,
160
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "3253e1f2-172e-4af4-a492-3b9c6e9e4797",
"operator": {
"name": "filter.operator.equals",
"type": "string",
"operation": "equals"
},
"leftValue": "={{ $json.error }}",
"rightValue": "Error in sub-node Switch Model"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "4a259078-aa74-4725-9e91-d2775bbd577f",
"name": "Loop finished without results",
"type": "n8n-nodes-base.set",
"position": [
1260,
60
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "b352627d-d692-47f8-8f8c-885b68073843",
"name": "output",
"type": "string",
"value": "The loop finished without a satisfying result"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "3b527ed3-a700-403d-8e3c-d0d55a83c9ea",
"name": "Unexpected error",
"type": "n8n-nodes-base.set",
"position": [
1260,
260
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "b352627d-d692-47f8-8f8c-885b68073843",
"name": "output",
"type": "string",
"value": "An unexpected error happened"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "2a48a244-25ab-4330-9e89-3f8a52b7fd0a",
"name": "Return result",
"type": "n8n-nodes-base.set",
"position": [
1420,
-460
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "b352627d-d692-47f8-8f8c-885b68073843",
"name": "output",
"type": "string",
"value": "={{ $json.text || $json.output }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "79da2795-800a-423d-ad5b-ec3b0498a5e6",
"name": "OpenAI 4o-mini",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
460,
580
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "c5884632-4f21-4e1e-a86d-77e3b18119b9",
"name": "OpenAI 4o",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
640,
580
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o",
"cachedResultName": "gpt-4o"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "0693ac6a-fd1e-4a1f-b7be-bd4a1021b6c1",
"name": "OpenAI o1",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
820,
580
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "o1",
"cachedResultName": "o1"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "f9fa467a-804d-4abf-84e3-06a88f9142b4",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1100,
-100
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "7c6bf364-1844-484f-8a1c-1ff87286c686",
"name": "Validate response",
"type": "@n8n/n8n-nodes-langchain.sentimentAnalysis",
"position": [
1040,
-300
],
"parameters": {
"options": {
"categories": "pass, fail",
"systemPromptTemplate": "You are a highly intelligent and accurate sentiment analyzer. Analyze the sentiment of the provided text. Categorize it into one of the following: {categories}. Use the provided formatting instructions. Only output the JSON.\n\n> Evaluate the following customer support response. Give a short JSON answer with a field \u201cquality\u201d: \u201cpass\u201d or \u201cfail\u201d. Only return \u201cpass\u201d if the response:\n\n1. Acknowledges both the broken keyboard and the late delivery \n2. Uses a polite and empathetic tone \n3. Offers a clear resolution or next step (like refund, replacement, or contact support)"
},
"inputText": "={{ $json.text }}"
},
"typeVersion": 1
},
{
"id": "a7be0179-e246-4f75-8863-d03eefe9d8ac",
"name": "Generate response",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"onError": "continueErrorOutput",
"position": [
660,
80
],
"parameters": {
"text": "={{ $('When chat message received').item.json.chatInput }}",
"messages": {
"messageValues": [
{
"message": "=You\u2019re an AI assistant replying to a customer who is upset about a faulty product and late delivery. The customer uses sarcasm and is vague. Write a short, polite response, offering help."
}
]
},
"promptType": "define"
},
"retryOnFail": false,
"typeVersion": 1.6
},
{
"id": "273f4025-2aeb-4a67-859a-690a3a086f82",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
380,
-160
],
"parameters": {
"width": 480,
"height": 140,
"content": "### Customer complaint - example\n\nI really *love* waiting two weeks just to get a keyboard that doesn\u2019t even work. Great job. Any chance I could actually use the thing I paid for sometime this month?"
},
"typeVersion": 1
},
{
"id": "a7806fab-fdc2-4feb-be53-fcea81ede105",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
380,
0
],
"parameters": {
"color": 7,
"width": 220,
"height": 240,
"content": "Defines the LLM node by index which should be used."
},
"typeVersion": 1
},
{
"id": "0117d8d8-672e-458a-a9dd-30b50e05f343",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
480,
240
],
"parameters": {
"color": 7,
"width": 380,
"height": 200,
"content": "Dynamically connects the LLM by the index provided in the previous node."
},
"typeVersion": 1
},
{
"id": "66066bad-4fd3-4e68-88bb-0b95fd9a6e49",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
980,
60
],
"parameters": {
"color": 7,
"width": 220,
"height": 260,
"content": "Check if LangChain Code Node ran into error. _Currently only supports error output from main Node_"
},
"typeVersion": 1
},
{
"id": "b9101226-0035-4de3-8720-f783d13e0cca",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
600,
0
],
"parameters": {
"color": 7,
"width": 380,
"height": 240,
"content": "Generates a polite answer based on the customers complaint."
},
"typeVersion": 1
},
{
"id": "ee7d70ee-2eb7-494f-ad74-2cb6108ba0ed",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
980,
-360
],
"parameters": {
"color": 7,
"width": 380,
"height": 220,
"content": "Analyses the generated answer by certain criteria"
},
"typeVersion": 1
},
{
"id": "03bde6f5-27b1-4568-96fb-5ece77d7b2e5",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
1360,
-280
],
"parameters": {
"color": 7,
"width": 220,
"height": 240,
"content": "Increases the index to choose the next available LLM on the next run"
},
"typeVersion": 1
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "52381ffc-bdf4-4243-bc35-462dedb929bd",
"connections": {
"OpenAI 4o": {
"ai_languageModel": [
[
{
"node": "Switch Model",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"OpenAI o1": {
"ai_languageModel": [
[
{
"node": "Switch Model",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Switch Model": {
"ai_outputParser": [
[]
],
"ai_languageModel": [
[
{
"node": "Generate response",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Set LLM index": {
"main": [
[
{
"node": "Generate response",
"type": "main",
"index": 0
}
]
]
},
"OpenAI 4o-mini": {
"ai_languageModel": [
[
{
"node": "Switch Model",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Generate response": {
"main": [
[
{
"node": "Validate response",
"type": "main",
"index": 0
}
],
[
{
"node": "Check for expected error",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "Validate response",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Validate response": {
"main": [
[
{
"node": "Return result",
"type": "main",
"index": 0
}
],
[
{
"node": "Increase LLM index",
"type": "main",
"index": 0
}
]
]
},
"Increase LLM index": {
"main": [
[
{
"node": "No Operation, do nothing",
"type": "main",
"index": 0
}
]
]
},
"Check for expected error": {
"main": [
[
{
"node": "Loop finished without results",
"type": "main",
"index": 0
}
],
[
{
"node": "Unexpected error",
"type": "main",
"index": 0
}
]
]
},
"No Operation, do nothing": {
"main": [
[
{
"node": "Set LLM index",
"type": "main",
"index": 0
}
]
]
},
"When chat message received": {
"main": [
[
{
"node": "Set LLM index",
"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.
openAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
How this works
This workflow enables seamless switching between large language models during conversations, ensuring optimal responses by selecting the most suitable LLM based on context or user needs without interrupting the flow. It's ideal for developers building AI chatbots or virtual assistants that require flexibility across models like OpenAI's GPT series for varied tasks such as sentiment analysis or general querying. The key step involves a code node that dynamically routes inputs to the chosen LLM via the lmChatOpenAi integration, followed by sentiment analysis to refine interactions.
Use this when your application demands adaptive AI responses, such as in customer support bots handling multilingual queries or creative writing tools needing specialised models. Avoid it for simple, single-model setups where overhead from the 22-node chain adds unnecessary complexity. Common variations include adding more LLMs like Anthropic's Claude or integrating with external APIs for model selection based on cost or performance metrics.
About this workflow
Dynamically switch between LLMs Template. Uses chatTrigger, noOp, lmChatOpenAi, sentimentAnalysis. Chat trigger; 22 nodes.
Source: https://github.com/Zie619/n8n-workflows — 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.
Testing Mulitple Local LLM with LM Studio. Uses stickyNote, httpRequest, chatTrigger, dateTime. Chat trigger; 21 nodes.
HTTP Request, Chat Trigger, OpenAI Chat +2
Testing Mulitple Local LLM with LM Studio. Uses stickyNote, httpRequest, chatTrigger, dateTime. Chat trigger; 21 nodes.
HTTP Request, Chat Trigger, OpenAI Chat +2
🔍🛠️ Tavily Search & Extract - Template. Uses stickyNote, httpRequest, lmChatOpenAi, chainLlm. Chat trigger; 17 nodes.
HTTP Request, OpenAI Chat, Chain Llm +1
Force Ai To Use A Specific Output Format. Uses manualTrigger, chainLlm, outputParserStructured, outputParserAutofixing. Event-driven trigger; 11 nodes.
Chain Llm, Output Parser Structured, Output Parser Autofixing +1
This n8n workflow template enables querying Excel data stored in an Oracle Database using natural language powered by Oracle Select AI.
Oracle Database, Chat Trigger