How can I resolve the 400 error when uploading Airtable attachments to OpenAI in n8n?

I'm trying to fetch an attachment from Airtable and then upload it to OpenAI using n8n.

I fetch the binary file via an HTTP Request node, and the binary data (stored in the property "attachment_0") is passed to the OpenAI node configured to upload a file. However, I receive the error "Bad request - please check your parameters" with the following details:

{
“errorMessage”: “Bad request - please check your parameters”,
“errorDescription”: “The browser (or proxy) sent a request that this server could not understand.”,
“errorDetails”: {
“rawErrorMessage”: [
“400 - {"error":{"message":"The browser (or proxy) sent a request that this server could not understand.","type":"server_error","param":null,"code":null}}”
],
“httpCode”: “400”
},
“n8nDetails”: {
“nodeName”: “OpenAI1”,
“nodeType”: “@n8n/n8n-nodes-langchain.openAi”,
“nodeVersion”: 1.8,
“resource”: “file”,
“operation”: “upload”,
“itemIndex”: 0,
“time”: “10.2.2025, 17:45:49”,
“n8nVersion”: “1.76.1 (Self Hosted)”,
“binaryDataMode”: “default”,
“stackTrace”: [
“NodeApiError: Bad request - please check your parameters”,
" at ExecuteContext.requestWithAuthentication (/usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/node-execute-functions.js:1297:19)",
" at processTicksAndRejections (node:internal/process/task_queues:95:5)",
" at ExecuteContext.requestWithAuthentication (/usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/node-execute-functions.js:1480:20)",
" at ExecuteContext.apiRequest (/usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/nodes/vendors/OpenAi/transport/index.js:22:12)",
" at ExecuteContext.execute (/usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/nodes/vendors/OpenAi/actions/file/upload.operation.js:67:26)",
" at ExecuteContext.router (/usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/nodes/vendors/OpenAi/OpenAi.node.js:16:16)",
" at WorkflowExecute.runNode (/usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/execution-engine/workflow-execute.js:632:19)",
" at /usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/execution-engine/workflow-execute.js:890:51",
" at /usr/local/lib/node_modules/n8n/node_modules/n8n-core/dist/execution-engine/workflow-execute.js:1224:20"
]
}
}

The HTTP Request node appears to work correctly and returns a binary file, but the OpenAI upload isn’t processing the file as expected.

I’m running this workflow using the self-hosted Docker setup on my Synology NAS. Below is my workflow configuration:

JSONCopied
{
  "nodes": [
    {
      "parameters": {
        "pollTimes": {
          "item": [
            {
              "mode": "everyMinute"
            }
          ]
        },
        "authentication": "airtableTokenApi",
        "baseId": {
          "__rl": true,
          "value": "appQRMowuiI1tT0F6",
          "mode": "id"
        },
        "tableId": {
          "__rl": true,
          "value": "tblV197pgZaWa6Hz2",
          "mode": "id"
        },
        "triggerField": "Erstellt um",
        "additionalFields": {
          "fields": ""
        }
      },
      "type": "n8n-nodes-base.airtableTrigger",
      "typeVersion": 1,
      "position": [
        -1600,
        20
      ],
      "id": "577f5715-17c8-4d89-9588-60e091e43f61",
      "name": "Airtable Trigger",
      "credentials": {
        "airtableTokenApi": {
          "id": "QUrLt5QMFeVRiqWY",
          "name": "Airtable account"
        }
      }
    },
    {
      "parameters": {
        "url": "={{ $json.fields.Anhang[1].url }}",
        "options": {
          "response": {
            "response": {
              "responseFormat": "file",
              "outputPropertyName": "attachment_0"
            }
          }
        }
      },
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        -1380,
        20
      ],
      "id": "0506ec51-ba22-4681-9f2b-1a0492fa3857",
      "name": "HTTP Request1"
    },
    {
      "parameters": {
        "resource": "file",
        "binaryPropertyName": "attachment_0",
        "options": {}
      },
      "type": "@n8n/n8n-nodes-langchain.openAi",
      "typeVersion": 1.8,
      "position": [
        -1120,
        20
      ],
      "id": "f61111b7-1af5-4857-b4ed-d9cc8b85a189",
      "name": "OpenAI1",
      "credentials": {
        "openAiApi": {
          "id": "Xi5oQW7t2TS4QaVr",
          "name": "n8n"
        }
      }
    }
  ],
  "connections": {
    "Airtable Trigger": {
      "main": [
        [
          {
            "node": "HTTP Request1",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "HTTP Request1": {
      "main": [
        [
          {
            "node": "OpenAI1",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "pinData": {},
  "meta": {
    "templateCredsSetupCompleted": true,
    "instanceId": "2b8389b15f8febc60e4ea42193783d59be45929cefbab62df923cf5dfbb50b3e"
  }
}

I'm looking for guidance on how to ensure the binary file data is structured correctly for OpenAI's API. Could this be related to missing file properties like a proper MIME type or file name?

Any troubleshooting tips or insights would be greatly appreciated.