Requesty — Multi-model Chain (gpt-5.4 → Gemini 3.5 Flash → Claude Opus 4.7)

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 →

{
  "name": "Requesty \u2014 Multi-Model Chain (GPT-5.4 \u2192 Gemini 3.5 Flash \u2192 Claude Opus 4.7)",
  "nodes": [
    {
      "parameters": {},
      "id": "aaaaaaaa-0001-0001-0001-000000000001",
      "name": "When clicking Test workflow",
      "type": "n8n-nodes-base.manualTrigger",
      "typeVersion": 1,
      "position": [
        -200,
        300
      ]
    },
    {
      "parameters": {
        "promptType": "define",
        "text": "Write a short, punchy product tagline for a developer tool that routes LLM requests across many model providers. Reply with just the tagline, nothing else.",
        "options": {
          "systemMessage": "You are a creative copywriter. Be concise and original."
        }
      },
      "id": "bbbbbbbb-0002-0002-0002-000000000002",
      "name": "1) Draft (GPT-5.4)",
      "type": "@n8n/n8n-nodes-langchain.chainLlm",
      "typeVersion": 1.6,
      "position": [
        120,
        200
      ]
    },
    {
      "parameters": {
        "model": "openai-responses/gpt-5.4",
        "options": {}
      },
      "id": "cccccccc-0003-0003-0003-000000000003",
      "name": "Requesty (GPT-5.4)",
      "type": "@requesty/n8n-nodes-requesty.lmChatRequesty",
      "typeVersion": 1,
      "position": [
        120,
        420
      ]
    },
    {
      "parameters": {
        "promptType": "define",
        "text": "Here is a draft tagline from another model:\n\n\"{{ $json.text }}\"\n\nCritique it in 2-3 bullet points, then produce one improved version. Keep your whole reply under 80 words.",
        "options": {
          "systemMessage": "You are a sharp marketing editor. You refine other people's drafts."
        }
      },
      "id": "dddddddd-0004-0004-0004-000000000004",
      "name": "2) Refine (Gemini 3.5 Flash)",
      "type": "@n8n/n8n-nodes-langchain.chainLlm",
      "typeVersion": 1.6,
      "position": [
        560,
        200
      ]
    },
    {
      "parameters": {
        "model": "vertex/gemini-3.5-flash",
        "options": {}
      },
      "id": "eeeeeeee-0005-0005-0005-000000000005",
      "name": "Requesty (Gemini 3.5 Flash)",
      "type": "@requesty/n8n-nodes-requesty.lmChatRequesty",
      "typeVersion": 1,
      "position": [
        560,
        420
      ]
    },
    {
      "parameters": {
        "promptType": "define",
        "text": "Two earlier models worked on a product tagline. Here is the latest critique-and-rewrite:\n\n{{ $json.text }}\n\nMake the final call: choose or write the single best tagline, then explain in one sentence why it wins. Format your reply as:\n\nFINAL: <tagline>\nWHY: <one sentence>",
        "options": {
          "systemMessage": "You are the senior creative director. You make the final decision and stand behind it."
        }
      },
      "id": "ffffffff-0006-0006-0006-000000000006",
      "name": "3) Finalize (Claude Opus 4.7)",
      "type": "@n8n/n8n-nodes-langchain.chainLlm",
      "typeVersion": 1.6,
      "position": [
        1000,
        200
      ]
    },
    {
      "parameters": {
        "model": "anthropic/claude-opus-4-7",
        "options": {}
      },
      "id": "11110007-0007-0007-0007-000000000007",
      "name": "Requesty (Claude Opus 4.7)",
      "type": "@requesty/n8n-nodes-requesty.lmChatRequesty",
      "typeVersion": 1,
      "position": [
        1000,
        420
      ]
    }
  ],
  "connections": {
    "When clicking Test workflow": {
      "main": [
        [
          {
            "node": "1) Draft (GPT-5.4)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "1) Draft (GPT-5.4)": {
      "main": [
        [
          {
            "node": "2) Refine (Gemini 3.5 Flash)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "2) Refine (Gemini 3.5 Flash)": {
      "main": [
        [
          {
            "node": "3) Finalize (Claude Opus 4.7)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Requesty (GPT-5.4)": {
      "ai_languageModel": [
        [
          {
            "node": "1) Draft (GPT-5.4)",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "Requesty (Gemini 3.5 Flash)": {
      "ai_languageModel": [
        [
          {
            "node": "2) Refine (Gemini 3.5 Flash)",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "Requesty (Claude Opus 4.7)": {
      "ai_languageModel": [
        [
          {
            "node": "3) Finalize (Claude Opus 4.7)",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1"
  },
  "meta": {
    "templateId": "requesty-multi-model-chain"
  }
}

About this workflow

Requesty — Multi-Model Chain (GPT-5.4 → Gemini 3.5 Flash → Claude Opus 4.7). Uses chainLlm, @requesty/n8n-nodes-requesty. Event-driven trigger; 7 nodes.

Source: https://github.com/requestyai/n8n-requesty/blob/main/test-workflow-chain.json — original creator credit. Request a take-down →

More AI & RAG workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.