Utiliser le modèle GPT-Image-1 d'OpenAI pour générer et modifier des images

• •Peut nécessiter les informations d'identification d'authentification de l'API cible

{
  "meta": {
    "instanceId": "96b96d0aa1e4ff5d5b6779332b149e3ef3364191562d79083d0309cf3ddfa53e"
  },
  "nodes": [
    {
      "id": "0e78a29e-87ba-4665-84c1-a413c45e25dc",
      "name": "Lors du clic sur 'Tester le workflow'",
      "type": "n8n-nodes-base.manualTrigger",
      "position": [
        -420,
        -40
      ],
      "parameters": {},
      "typeVersion": 1
    },
    {
      "id": "fe7b054a-e0c9-4642-a97f-6bec60a1ec55",
      "name": "Éditer l'image (OpenAI)",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        500,
        -40
      ],
      "parameters": {
        "url": "https://api.openai.com/v1/images/edits",
        "method": "POST",
        "options": {},
        "sendBody": true,
        "contentType": "multipart-form-data",
        "authentication": "predefinedCredentialType",
        "bodyParameters": {
          "parameters": [
            {
              "name": "image",
              "parameterType": "formBinaryData",
              "inputDataFieldName": "data"
            },
            {
              "name": "prompt",
              "value": "add a mask with horns"
            },
            {
              "name": "model",
              "value": "gpt-image-1"
            },
            {
              "name": "n",
              "value": "1"
            },
            {
              "name": "size",
              "value": "1024x1024"
            },
            {
              "name": "quality",
              "value": "high"
            }
          ]
        },
        "nodeCredentialType": "openAiApi"
      },
      "credentials": {
        "openAiApi": {
          "id": "JyI0PkPec1FrpMkt",
          "name": "OpenAi AIFB account"
        }
      },
      "typeVersion": 4.2
    },
    {
      "id": "1e1df05c-d8f9-4033-87ee-70be344ab961",
      "name": "Appel de création d'image",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        -120,
        -40
      ],
      "parameters": {
        "url": "https://api.openai.com/v1/images/generations",
        "method": "POST",
        "options": {},
        "sendBody": true,
        "authentication": "predefinedCredentialType",
        "bodyParameters": {
          "parameters": [
            {
              "name": "model",
              "value": "gpt-image-1"
            },
            {
              "name": "prompt",
              "value": "A cute red panda like dark super hero"
            },
            {
              "name": "n",
              "value": "={{Number(1)}}"
            },
            {
              "name": "size",
              "value": "1024x1024"
            },
            {
              "name": "moderation",
              "value": "low"
            },
            {
              "name": "background",
              "value": "auto"
            }
          ]
        },
        "nodeCredentialType": "openAiApi"
      },
      "credentials": {
        "openAiApi": {
          "id": "JyI0PkPec1FrpMkt",
          "name": "OpenAi AIFB account"
        }
      },
      "typeVersion": 4.2
    },
    {
      "id": "4c44da91-0d12-4e7f-bc89-5accddd837d7",
      "name": "Convertir le binaire json en fichier",
      "type": "n8n-nodes-base.convertToFile",
      "position": [
        200,
        -40
      ],
      "parameters": {
        "options": {
          "fileName": "name_example",
          "mimeType": "image/png"
        },
        "operation": "toBinary",
        "sourceProperty": "data[0].b64_json"
      },
      "typeVersion": 1.1
    },
    {
      "id": "3b8936b7-f0a2-4776-b10a-f06ceb9af31d",
      "name": "Convertir le binaire json en fichier final",
      "type": "n8n-nodes-base.convertToFile",
      "position": [
        820,
        -40
      ],
      "parameters": {
        "options": {
          "fileName": "",
          "mimeType": "image/png"
        },
        "operation": "toBinary",
        "sourceProperty": "data[0].b64_json"
      },
      "typeVersion": 1.1
    },
    {
      "id": "3d3238d5-6040-4b74-8e6a-9e1e64198099",
      "name": "Note adhésive",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -500,
        -200
      ],
      "parameters": {
        "height": 320,
        "content": "### 🧪 Manual Trigger\nStarts the workflow manually. Ideal for testing and debugging purposes.\n"
      },
      "typeVersion": 1
    },
    {
      "id": "c3378100-f688-4199-a038-83b9220afa91",
      "name": "Note adhésive 1",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -200,
        -320
      ],
      "parameters": {
        "color": 3,
        "width": 280,
        "height": 440,
        "content": "### 🎨 Image Generation (OpenAI)\nSends a POST request to the OpenAI `/v1/images/generations` endpoint.\n\n- Uses `gpt-image-1` model\n- Generates an image from a given prompt\n- Returns a base64-encoded image (`b64_json`)\n\n📌 Output: `data[0].b64_json`\n"
      },
      "typeVersion": 1
    },
    {
      "id": "82a880de-74de-44e5-8448-f487c9376d0e",
      "name": "Note adhésive 2",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        100,
        -200
      ],
      "parameters": {
        "color": 5,
        "width": 280,
        "height": 320,
        "content": "### 🧾 Convert base64 to File\nConverts the `b64_json` field into a binary PNG file to use in the next step.\n\n📤 Output: Binary image under the `data` field\n"
      },
      "typeVersion": 1
    },
    {
      "id": "42ccb29f-b820-4791-9683-4eb0f00ff2d3",
      "name": "Note adhésive 3",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        420,
        -320
      ],
      "parameters": {
        "color": 3,
        "width": 280,
        "height": 440,
        "content": "### ✏️ Image Editing (OpenAI)\nSends the binary image to OpenAI’s `/v1/images/edits` endpoint with a descriptive prompt.\n\n- Model: `gpt-image-1`\n- Format: `multipart/form-data`\n- Requires a real file (not base64)\n- Supports optional `mask` input\n\n📥 Input: Binary image from `data`\n📤 Output: Edited image in `b64_json`\n"
      },
      "typeVersion": 1
    },
    {
      "id": "4c8846ab-b3f2-4c7c-9283-5a40a55b816d",
      "name": "Note adhésive 4",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        740,
        -240
      ],
      "parameters": {
        "color": 5,
        "width": 280,
        "height": 360,
        "content": "### 🧾 Final Conversion (base64 → File)\nConverts the edited image (`b64_json`) into a downloadable or previewable PNG file.\n\n📤 Output: Final binary image\n"
      },
      "typeVersion": 1
    },
    {
      "id": "2b2533f8-b7aa-4499-970e-9b0546b73c0e",
      "name": "Note adhésive 5",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1240,
        -860
      ],
      "parameters": {
        "color": 6,
        "width": 700,
        "height": 980,
        "content": "## 🧠 Image AI Workflow Overview\n\nThis workflow uses OpenAI's image generation and editing APIs with the `gpt-image-1` model.\n\n### 🔑 Requirements:\n- You **must use your own OpenAI API key** from https://platform.openai.com/account/api-keys\n- Create a credential in n8n called `OpenAi AIFB account` (or use your own name)\n\n---\n\n### 💰 Cost Warning:\n- This model is **powerful but expensive**.\n- Each image costs **$0.020 to $0.190** depending on resolution and type.\n- Always monitor your usage via the [OpenAI dashboard](https://platform.openai.com/account/usage)\n\n---\n\n### 🔍 Why use `gpt-image-1`?\n- Unmatched **semantic control**: you can edit specific parts of images with detailed prompts.\n- Supports **multiple input images**, coherent edits, and future multi-modal tasks.\n- Editing works with or without a transparency `mask`.\n\n---\n\n### 🔧 Suggested Nodes to Expand Workflow:\n- **Webhook** (trigger via your frontend or app)\n- **Telegram / Slack** (prompt image generation from chat)\n- **Set node** (inject dynamic prompts or user context)\n- **IF / Switch** (change behavior depending on prompt type)\n- **Merge** (combine multiple image generations)\n- **HTTP Request** (send generated images to external APIs or CMS)\n\n---\n\n### 💡 Example Use Cases:\n- Marketing teams: generate product visuals on demand\n- Designers: edit and re-style illustrations without Photoshop\n- E-commerce: dynamic generation of themed mockups\n- Content creators: create blog and social thumbnails in bulk\n\n---\n\n> ⚠️ Don't forget to add rate limiting or batch controls if generating large volumes!\n"
      },
      "typeVersion": 1
    }
  ],
  "pinData": {},
  "connections": {
    "1e1df05c-d8f9-4033-87ee-70be344ab961": {
      "main": [
        [
          {
            "node": "4c44da91-0d12-4e7f-bc89-5accddd837d7",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "fe7b054a-e0c9-4642-a97f-6bec60a1ec55": {
      "main": [
        [
          {
            "node": "3b8936b7-f0a2-4776-b10a-f06ceb9af31d",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "4c44da91-0d12-4e7f-bc89-5accddd837d7": {
      "main": [
        [
          {
            "node": "fe7b054a-e0c9-4642-a97f-6bec60a1ec55",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "0e78a29e-87ba-4665-84c1-a413c45e25dc": {
      "main": [
        [
          {
            "node": "1e1df05c-d8f9-4033-87ee-70be344ab961",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}