Troubleshooting 400 Bad Request Error When Using 1min.AI API in n8n

Below is a summary of the issue you're experiencing:

Summary of the Problem

You're integrating the 1min.AI API into n8n using the HTTP Request node. Even though you’re following the official API documentation and sending a static JSON body that looks like this:

JSONCopied
{
  "type": "CHAT_WITH_AI",
  "model": "gpt-4o-mini",
  "promptObject": {
    "prompt": "hi",
    "isMixed": false,
    "imageList": [],
    "webSearch": true,
    "numOfSite": 1,
    "maxWord": 500
  }
}

The API keeps responding with a 400 Bad Request error. The error message indicates that the required fields are either missing or not recognized properly:

"type should not be empty, type must be a string, model should not be empty, model must be a string"

What You've Tried

• Sending a minimal request: Reduced the payload to only the required fields (type, model, and promptObject.prompt), but the error persists.
• Validating the API key: Confirmed the API key works in Postman.
• Switching between JSON and RAW modes: Both modes result in the same error.
• Using jsonParameters=true: Ensured this n8n setting is enabled, yet no improvement was seen.

Implications of the Error Message

The error strongly suggests that even though your JSON body is properly structured, the API is not receiving the expected string values for both the "type" and "model" fields. In other words, although your request appears correct, there might be an issue downstream in how n8n is transmitting or formatting the payload.

Potential Causes to Consider

• Content-Type Header: Make sure that the HTTP Request node is sending the payload with the Content-Type: application/json header. An incorrect header might cause the destination API to misinterpret the body content.
• Payload Formatting: Sometimes, even if the JSON looks correct in the manual input, the node configuration might be altering the request format or encoding before sending it. Double-check that no unexpected wrapping or modifications occur in the HTTP Request node.
• n8n Node Settings: Although you've verified the jsonParameters setting, inspect any additional options (like additional query parameters or any node overrides) that could affect how the body is sent.

Next Steps

1. Verify Headers: Confirm the HTTP Request node sends the appropriate headers including Content-Type: application/json.
2. Log the Outgoing Request: If possible, capture the raw outgoing request (using tools such as a proxy or network monitor) to inspect the exact payload being sent.
3. Test with a Simplified Request: Try sending a simpler payload (even if it diverges from the API example) to see if altering the structure slightly makes a difference.
4. Compare with Postman Setup: Since the key works in Postman, compare the HTTP request details (headers, body formatting) between Postman and n8n.

By checking these points, you should be able to pinpoint whether the issue lies in the way n8n is packaging the request or elsewhere in the API interaction.

Keep in mind, this behavior is similar to a known common issue in the HTTP Request node when parameters and body formats aren’t exactly what the API expects.