This workflow follows the Agent → Chainllm 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": "43gMd18arOcxqDcC",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "LLM Chaining examples",
"tags": [],
"nodes": [
{
"id": "35e53ce7-06b4-47ca-b7f3-b147bd059fcf",
"name": "When clicking \u2018Test workflow\u2019",
"type": "n8n-nodes-base.manualTrigger",
"position": [
200,
520
],
"parameters": {},
"typeVersion": 1
},
{
"id": "aeef734e-1c3b-4a91-93ae-2ae9c50951b8",
"name": "HTTP Request",
"type": "n8n-nodes-base.httpRequest",
"position": [
400,
520
],
"parameters": {
"url": "https://blog.n8n.io/",
"options": {}
},
"typeVersion": 4.2
},
{
"id": "7f6b95eb-df8c-4f0f-ba69-6b298d624ccd",
"name": "Markdown",
"type": "n8n-nodes-base.markdown",
"position": [
600,
520
],
"parameters": {
"html": "={{ $json.data }}",
"options": {},
"destinationKey": "markdown"
},
"typeVersion": 1
},
{
"id": "994dbe06-4c25-4fb3-a8f3-566eb5b66c6d",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
160,
340
],
"parameters": {
"color": 4,
"width": 700,
"height": 360,
"content": "# Connect to one of the blue sections -->\n## This can be anything:\n- Chat input\n- Trigger from external system\n- CRON-scheduled event"
},
"typeVersion": 1
},
{
"id": "8ba3039d-dabf-43b7-ab35-117332f65ced",
"name": "Anthropic Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
1460,
-20
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "claude-3-7-sonnet-20250219",
"cachedResultName": "Claude 3.7 Sonnet"
},
"options": {
"temperature": 0.5
}
},
"credentials": {
"anthropicApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "7e1da020-e01d-410c-aa7f-a19d6e1c368d",
"name": "Anthropic Chat Model1",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
1820,
-20
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "claude-3-7-sonnet-20250219",
"cachedResultName": "Claude 3.7 Sonnet"
},
"options": {
"temperature": 0.5
}
},
"credentials": {
"anthropicApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "620503cb-2d51-4102-8975-75255cf15b1b",
"name": "Anthropic Chat Model2",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
2180,
-20
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "claude-3-7-sonnet-20250219",
"cachedResultName": "Claude 3.7 Sonnet"
},
"options": {
"temperature": 0.5
}
},
"credentials": {
"anthropicApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "5f0d11ce-c1ea-4c36-8b2d-d3f70b19f0ba",
"name": "Anthropic Chat Model3",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
2540,
-20
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "claude-3-7-sonnet-20250219",
"cachedResultName": "Claude 3.7 Sonnet"
},
"options": {
"temperature": 0.5
}
},
"credentials": {
"anthropicApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "f973d01e-fad7-4143-8379-54438f5412cb",
"name": "Merge",
"type": "n8n-nodes-base.merge",
"position": [
2440,
360
],
"parameters": {
"mode": "combine",
"options": {
"includeUnpaired": true
},
"combineBy": "combineByPosition"
},
"typeVersion": 3.1
},
{
"id": "c7e58b90-bc96-421c-88f2-4e9f95f87248",
"name": "Simple Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
2680,
780
],
"parameters": {
"sessionKey": "fixed_session",
"sessionIdType": "customKey",
"contextWindowLength": 10
},
"typeVersion": 1.3
},
{
"id": "0e606f7c-2cdb-4e34-8c0b-2303996077fb",
"name": "Clean memory",
"type": "@n8n/n8n-nodes-langchain.memoryManager",
"position": [
1500,
480
],
"parameters": {
"mode": "delete",
"deleteMode": "all"
},
"typeVersion": 1.1
},
{
"id": "af0fb574-9964-4f7d-8348-a2cf614b8562",
"name": "Initial prompts",
"type": "n8n-nodes-base.set",
"position": [
1880,
480
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "84a50c9c-2265-4dd6-a774-efc852615862",
"name": "system_prompt",
"type": "string",
"value": "You are a helpful assistant"
},
{
"id": "559f19f7-042c-465e-b85f-ab52cfbab04a",
"name": "step1",
"type": "string",
"value": "What is on this page?"
},
{
"id": "6791cd09-c5f7-48c8-b753-8d383db6863f",
"name": "step2",
"type": "string",
"value": "List all authors on this page"
},
{
"id": "1f92ac04-e5dd-4161-afde-14562aea454c",
"name": "step3",
"type": "string",
"value": "List all posts on this page"
},
{
"id": "ad8ee0b0-fa7d-4f4a-85a8-82d0d0dc0a40",
"name": "step4",
"type": "string",
"value": "Make a bold funny joke based on the content on this page"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "6743e44a-cc76-4e73-b4f3-ba7c65d179d3",
"name": "Split Out",
"type": "n8n-nodes-base.splitOut",
"position": [
2240,
480
],
"parameters": {
"include": "selectedOtherFields",
"options": {},
"fieldToSplitOut": "data",
"fieldsToInclude": "system_prompt"
},
"typeVersion": 1
},
{
"id": "caddd26c-ee84-455f-8ee6-aecf21536930",
"name": "Reshape",
"type": "n8n-nodes-base.set",
"position": [
2060,
480
],
"parameters": {
"mode": "raw",
"include": "selected",
"options": {},
"jsonOutput": "={ \"data\" : {{ Object.entries($json).filter(([key]) => key !== \"system_prompt\").map(([key, value]) => ({ step: key, instruction: value })) }}\n}",
"includeFields": "system_prompt",
"includeOtherFields": true
},
"typeVersion": 3.4
},
{
"id": "bd244988-d074-42f3-af42-960e5aa1d35d",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
1840,
400
],
"parameters": {
"width": 540,
"height": 240,
"content": "# An array of prompts here"
},
"typeVersion": 1
},
{
"id": "7e9e5287-8d4e-43a9-b8cf-ae26a177bfbb",
"name": "Anthropic Chat Model4",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
2600,
600
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "claude-3-7-sonnet-20250219",
"cachedResultName": "Claude 3.7 Sonnet"
},
"options": {
"thinking": false,
"temperature": 0.5
}
},
"credentials": {
"anthropicApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "47816a45-b906-47ef-9510-c63867bfc8b7",
"name": "Merge2",
"type": "n8n-nodes-base.merge",
"position": [
1860,
1120
],
"parameters": {
"mode": "combine",
"options": {},
"combineBy": "combineAll"
},
"typeVersion": 3
},
{
"id": "e63b89a1-c2ca-4ed2-ae50-e3a7b429609c",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
2040,
1020
],
"parameters": {
"width": 320,
"height": 520,
"content": "## Make sure URL matches\n### \u26a0\ufe0f Cloud users!\nReplace `{{ $env.WEBHOOK_URL }}` \nwith your n8n instance URL"
},
"typeVersion": 1
},
{
"id": "7b99df1a-bf6c-4cf1-b58a-346873136715",
"name": "Basic LLM Chain4",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
2440,
1300
],
"parameters": {
"text": "={{ $json.body.userprompt }}\n\nHere's page data:\n~~~~\n{{ $json.body.markdown }}\n~~~~",
"promptType": "define"
},
"typeVersion": 1.5
},
{
"id": "6f6e0667-5164-4b65-a796-1d2112c7c072",
"name": "Split Out1",
"type": "n8n-nodes-base.splitOut",
"position": [
1680,
1340
],
"parameters": {
"options": {},
"fieldToSplitOut": "userprompt"
},
"typeVersion": 1
},
{
"id": "9dfd2145-2427-4131-92d2-99aca620217f",
"name": "Anthropic Chat Model5",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
2420,
1460
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "claude-3-7-sonnet-20250219",
"cachedResultName": "Claude 3.7 Sonnet"
},
"options": {
"thinking": false,
"temperature": 0.5
}
},
"credentials": {
"anthropicApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "616fc635-107d-4929-b9d6-4ccd34e42909",
"name": "Webhook",
"type": "n8n-nodes-base.webhook",
"position": [
2140,
1400
],
"parameters": {
"path": "58d2b899-e09c-45bf-b59b-961a5d7a2470",
"options": {},
"httpMethod": "POST",
"responseMode": "lastNode"
},
"typeVersion": 2
},
{
"id": "c863252b-f8b6-4704-be4e-a69d3005a85a",
"name": "CONNECT ME",
"type": "n8n-nodes-base.noOp",
"position": [
1240,
-220
],
"parameters": {},
"typeVersion": 1
},
{
"id": "90ab4402-cbea-4441-9097-558ec72e5d38",
"name": "CONNECT ME1",
"type": "n8n-nodes-base.noOp",
"position": [
1280,
340
],
"parameters": {},
"typeVersion": 1
},
{
"id": "1c04650f-8043-496f-aeab-866e85548f9d",
"name": "CONNECT ME2",
"type": "n8n-nodes-base.noOp",
"position": [
1280,
1100
],
"parameters": {},
"typeVersion": 1
},
{
"id": "4097f12d-eba7-477a-9152-da5eb8c9aa03",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
960,
-300
],
"parameters": {
"color": 5,
"width": 1980,
"height": 440,
"content": "# 1 - Naive Chaining\n### PROs:\n- Easy to setup\n- Beginner-friendly\n\n### CONs\n- Not scalable\n- Hard to maintain long chains\n- SLOOOW!"
},
"typeVersion": 1
},
{
"id": "ce806bc6-a57e-47da-bbba-4698c3956022",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
960,
240
],
"parameters": {
"color": 5,
"width": 2160,
"height": 660,
"content": "# 2 - Iterative Agent Processing\n\n### PROs:\n- Scalable\n- All inputs & outputs in a single node\n- Supports Agent memory\n\n### CONs\n- Still Slow!"
},
"typeVersion": 1
},
{
"id": "49c4507f-de1e-422b-8058-db82668614d3",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
960,
1000
],
"parameters": {
"color": 5,
"width": 1880,
"height": 600,
"content": "# 3 - Parallel Processing\n\n### PROs:\n- Scalable\n- All inputs & outputs in a single place\n- FAST!\n\n### CONs\n- Independent requests\n (no Agent memory)"
},
"typeVersion": 1
},
{
"id": "c30b8132-9291-4855-89ec-6a98bcee8247",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
1420,
1260
],
"parameters": {
"width": 400,
"height": 240,
"content": "# Array of prompts here"
},
"typeVersion": 1
},
{
"id": "4c1b5816-7393-47f6-8a88-008d8deea119",
"name": "Initial prompts1",
"type": "n8n-nodes-base.set",
"position": [
1460,
1340
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "ed7f1cc6-99d3-481c-b5fb-d9900d6ee0f6",
"name": "userprompt",
"type": "array",
"value": "=[\n\"What is on this page?\",\n\"List all authors on this page\",\n\"List all posts on this page\",\n\"Make a bold funny joke based on the content on this page\"\n]"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "8248a20f-1f90-42b0-8167-7ddcc90242a2",
"name": "LLM Chain - Step 1",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
1500,
-220
],
"parameters": {
"text": "={{ $('Markdown').first().json.markdown }}",
"messages": {
"messageValues": [
{
"message": "What is on this page?"
}
]
},
"promptType": "define"
},
"typeVersion": 1.6
},
{
"id": "3788b23b-ccdc-4326-8ce0-1e57934d23bd",
"name": "LLM Chain - Step 2",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
1860,
-220
],
"parameters": {
"text": "={{ $('Markdown').first().json.markdown }}",
"messages": {
"messageValues": [
{
"message": "List all authors on this page"
}
]
},
"promptType": "define"
},
"typeVersion": 1.6
},
{
"id": "89e69a39-bf13-4599-8ddc-a01c4590fb9c",
"name": "LLM Chain - Step 3",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
2220,
-220
],
"parameters": {
"text": "={{ $('Markdown').first().json.markdown }}",
"messages": {
"messageValues": [
{
"message": "List all posts on this page"
}
]
},
"promptType": "define"
},
"typeVersion": 1.6
},
{
"id": "7e395991-9404-490e-8946-0da8f81e7243",
"name": "LLM Chain - Step 4",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
2580,
-220
],
"parameters": {
"text": "={{ $('Markdown').first().json.markdown }}",
"messages": {
"messageValues": [
{
"message": "Make a bold funny joke based on the content on this page"
}
]
},
"promptType": "define"
},
"typeVersion": 1.6
},
{
"id": "efb8d836-8a4a-4a70-baed-4a9b77461aca",
"name": "All LLM steps here - sequentially",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
2640,
440
],
"parameters": {
"text": "={{ $json.markdown || \"\" }}\n{{ `Your task: ${$json.data.step}. ${$json.data.instruction}` }}",
"options": {
"systemMessage": "={{ $json.system_prompt }}"
},
"promptType": "define"
},
"typeVersion": 1.8
},
{
"id": "926b1705-a24c-4659-bf61-8ed97ade7290",
"name": "LLM steps - parallel",
"type": "n8n-nodes-base.httpRequest",
"position": [
2140,
1240
],
"parameters": {
"url": "={{ $env.WEBHOOK_URL }}webhook/58d2b899-e09c-45bf-b59b-961a5d7a2470",
"method": "POST",
"options": {},
"jsonBody": "={{ $json }}",
"sendBody": true,
"specifyBody": "json"
},
"typeVersion": 4.2
},
{
"id": "7748574b-1abd-4697-9644-db8bb79fb08d",
"name": "Merge output with initial prompts",
"type": "n8n-nodes-base.merge",
"position": [
2440,
1140
],
"parameters": {
"mode": "combine",
"options": {},
"combineBy": "combineByPosition"
},
"typeVersion": 3.1
},
{
"id": "b207d83b-ecda-4a9f-af78-cfbb2253c119",
"name": "Merge output with initial prompts1",
"type": "n8n-nodes-base.merge",
"position": [
3000,
380
],
"parameters": {
"mode": "combine",
"options": {},
"combineBy": "combineByPosition"
},
"typeVersion": 3.1
}
],
"active": true,
"settings": {
"callerPolicy": "workflowsFromSameOwner",
"executionOrder": "v1",
"saveDataSuccessExecution": "all"
},
"versionId": "7b1849db-1c4c-4943-89b1-184926649776",
"connections": {
"Merge": {
"main": [
[
{
"node": "All LLM steps here - sequentially",
"type": "main",
"index": 0
},
{
"node": "Merge output with initial prompts1",
"type": "main",
"index": 0
}
]
]
},
"Merge2": {
"main": [
[
{
"node": "LLM steps - parallel",
"type": "main",
"index": 0
},
{
"node": "Merge output with initial prompts",
"type": "main",
"index": 0
}
]
]
},
"Reshape": {
"main": [
[
{
"node": "Split Out",
"type": "main",
"index": 0
}
]
]
},
"Webhook": {
"main": [
[
{
"node": "Basic LLM Chain4",
"type": "main",
"index": 0
}
]
]
},
"Markdown": {
"main": [
[]
]
},
"Split Out": {
"main": [
[
{
"node": "Merge",
"type": "main",
"index": 1
}
]
]
},
"CONNECT ME": {
"main": [
[
{
"node": "LLM Chain - Step 1",
"type": "main",
"index": 0
}
]
]
},
"Split Out1": {
"main": [
[
{
"node": "Merge2",
"type": "main",
"index": 1
}
]
]
},
"CONNECT ME1": {
"main": [
[
{
"node": "Clean memory",
"type": "main",
"index": 0
},
{
"node": "Merge",
"type": "main",
"index": 0
}
]
]
},
"CONNECT ME2": {
"main": [
[
{
"node": "Initial prompts1",
"type": "main",
"index": 0
},
{
"node": "Merge2",
"type": "main",
"index": 0
}
]
]
},
"Clean memory": {
"main": [
[
{
"node": "Initial prompts",
"type": "main",
"index": 0
}
]
]
},
"HTTP Request": {
"main": [
[
{
"node": "Markdown",
"type": "main",
"index": 0
}
]
]
},
"Simple Memory": {
"ai_memory": [
[
{
"node": "All LLM steps here - sequentially",
"type": "ai_memory",
"index": 0
},
{
"node": "Clean memory",
"type": "ai_memory",
"index": 0
}
]
]
},
"Initial prompts": {
"main": [
[
{
"node": "Reshape",
"type": "main",
"index": 0
}
]
]
},
"Initial prompts1": {
"main": [
[
{
"node": "Split Out1",
"type": "main",
"index": 0
}
]
]
},
"LLM Chain - Step 1": {
"main": [
[
{
"node": "LLM Chain - Step 2",
"type": "main",
"index": 0
}
]
]
},
"LLM Chain - Step 2": {
"main": [
[
{
"node": "LLM Chain - Step 3",
"type": "main",
"index": 0
}
]
]
},
"LLM Chain - Step 3": {
"main": [
[
{
"node": "LLM Chain - Step 4",
"type": "main",
"index": 0
}
]
]
},
"Anthropic Chat Model": {
"ai_languageModel": [
[
{
"node": "LLM Chain - Step 1",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"LLM steps - parallel": {
"main": [
[
{
"node": "Merge output with initial prompts",
"type": "main",
"index": 1
}
]
]
},
"Anthropic Chat Model1": {
"ai_languageModel": [
[
{
"node": "LLM Chain - Step 2",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Anthropic Chat Model2": {
"ai_languageModel": [
[
{
"node": "LLM Chain - Step 3",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Anthropic Chat Model3": {
"ai_languageModel": [
[
{
"node": "LLM Chain - Step 4",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Anthropic Chat Model4": {
"ai_languageModel": [
[
{
"node": "All LLM steps here - sequentially",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Anthropic Chat Model5": {
"ai_languageModel": [
[
{
"node": "Basic LLM Chain4",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"All LLM steps here - sequentially": {
"main": [
[
{
"node": "Merge output with initial prompts1",
"type": "main",
"index": 1
}
]
]
},
"When clicking \u2018Test workflow\u2019": {
"main": [
[
{
"node": "HTTP Request",
"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.
anthropicApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
How this works
This workflow demonstrates practical chaining of large language models to build sophisticated AI interactions, enabling you to create responsive systems that handle complex queries by passing outputs sequentially between models. It's ideal for developers and AI enthusiasts experimenting with multi-step reasoning, such as generating detailed responses from initial prompts refined through subsequent LLM calls using integrations like Anthropic's chat model and HTTP Request for external data. The key step involves the chainLlm and agent nodes, which orchestrate the flow by linking multiple Anthropic Chat Model instances to maintain context and enhance accuracy across 38 interconnected nodes.
Use this when prototyping event-driven AI applications that require layered processing, like automated content generation or query refinement triggered by user events. Avoid it for production environments without customisation, as the manual trigger and sticky notes suit learning rather than scalable deployment. Common variations include swapping Anthropic for other LLM providers or adding memoryBufferWindow for longer conversation histories.
About this workflow
LLM Chaining examples. Uses manualTrigger, httpRequest, markdown, stickyNote. Event-driven trigger; 38 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.
This workflow demonstrates three distinct approaches to chaining LLM operations using Claude 3.7 Sonnet. Connect to any section to experience the differences in implementation, performance, and capabi
HTTP Request, Anthropic Chat, Memory Buffer Window +3
Flexible and scalable chatbot template, designed mainly for Spanish conversations but capable of handling English and other languages. Integrates Google Gemini API for text and image generation, and T
Agent, Google Gemini Chat, Chain Llm +6
Template Carnaval - time instagram. Uses toolWorkflow, lmChatOpenAi, memoryBufferWindow, agent. Event-driven trigger; 56 nodes.
Tool Workflow, OpenAI Chat, Memory Buffer Window +10
Example. Uses agent, httpRequest, memoryBufferWindow, telegram. Event-driven trigger; 55 nodes.
Agent, HTTP Request, Memory Buffer Window +3
Sourcing_Agent_LinkedIn_validator_production. Uses executeWorkflowTrigger, chainLlm, mySql, httpRequest. Event-driven trigger; 51 nodes.
Execute Workflow Trigger, Chain Llm, MySQL +9