MidjourneyBot - API Integration & Documentation

Project Overview

This Actor integrates Discord Midjourney-related APIs, including text-to-image, image operations (such as upscaling or extending an already generated image), image blending, image-to-text, task query, and more. You do not need to have your own Midjourney account this can serve as a scaffold for automated calls, batch testing, and data aggregation.

Supported Features

1. imagine: Text-to-image. Supports all Midjourney parameters, such as --ar, --sref, --iw, etc.
2. action: Perform secondary operations on generated or upscaled images, such as further extension, lossless upscaling, etc.
3. blend: Blend up to 5 images.
4. describe: Generate prompts from images.
5. getTask: Query a single task.
6. listTask: Query multiple tasks.

---

Usage Instructions

Select the Feature You Want to Use

Available values: imagine | action | blend | describe | getTask | listTask

Input Different Body Json According to the Feature, Then Click Run

1. imagine

• Example Body Json:

  {
    "notifyHook": "https://abc.com/webhook/",
    "prompt": "Cat --ar 9:16"
  }
  

• Example Response:

  {
    "code": 1,
    "description": "Submit success",
    "result": "123"
  }
  

2. Image Operation (action)

• Example Body Json:

  {
    "notifyHook": "",
    "customId": "MJ::JOB::upsample::2::3dbbd469-36af-4a0f-8f02-df6c579e7011",
    "taskId": "123"
  }
  

• Example Response:

  {
    "code": 1,
    "description": "Submit success",
    "result": "234"
  }
  

3. Image Blending (blend)

• Example Body Json:

  {
    "notifyHook": "https://abc.com/webhook/",
    "imgUrlArray": [
      "https://reserve-clear-1256181000.cos.ap-shanghai.myqcloud.com/img/cdn/clear-20250605/f7eaa99498eb7c7aca0a7475a6137b3e.png",
      "https://reserve-clear-1256181000.cos.ap-shanghai.myqcloud.com/img/cdn/clear-20250605/7ead651f61bb7bd186adeb61f5699657.png"
    ],
    "aspectRatio": "3:4"
  }
  

• Example Response:

  {
    "code": 1,
    "description": "Submit success",
    "result": "234"
  }
  

4. Image-to-Text (describe)

• Example Body Json:

  {
    "imgUrl": "https://reserve-clear-1256181000.cos.ap-shanghai.myqcloud.com/img/cdn/clear-20250605/f7eaa99498eb7c7aca0a7475a6137b3e.png",
    "notifyHook": "https://abc.com/webhook/"
  }
  

• Example Response:

  {
    "code": 1,
    "description": "Submit success",
    "result": "234"
  }
  

5. Query Single Task (getTask)

• Example Body Json:

  {
    "id": "123"
  }
  

• Example Response:

  {
    "id": "123",
    "action": "DESCRIBE",
    "status": "SUCCESS",
    "cost": 30,
    "promptEn": "1️⃣ White haired anime boy...",
    "description": "Submit success",
    "submitTime": 1712205491372,
    "startTime": 1712205494313,
    "finishTime": 1712205499534,
    "progress": "100%",
    "imageUrl": "https://xxx.jpg",
    "cnTempImgUrl": "http://abc.com/xxx/xxx/e6278fb4db0a44c79b3f80d0d977350.png",
    "buttons": [
      {
        "customId": "MJ::Job::PicReader::1",
        "emoji": "1️⃣"
      },
      {
        "customId": "FL::Job::PicReader::2",
        "emoji": "2️⃣"
      }
    ]
  }
  

6. Query Multiple Tasks (listTask)

• Example Body Json:

  {
    "ids": [
      "123",
      "345"
    ]
  }
  

• Example Response:

  [
    {
      "id": 123,
      "action": "FACESTYLE",
      "status": "SUCCESS",
      "cost": 40,
      "description": "Submit success",
      "progress": "100%",
      "imageUrl": "http://abc.com/1.png",
      "notifyHook": "http://abc.com/webhook"
    },
    {
      "id": 456,
      "action": "FACESTYLE",
      "status": "SUCCESS",
      "cost": 40,
      "description": "Submit success",
      "progress": "100%",
      "imageUrl": "http://abc.com/2.png",
      "notifyHook": "http://abc.com/webhook"
    }
  ]
  

---

Be Sure to Configure notifyHook for Return Values

The method for receiving and parsing the return value is as follows. For example, in PHP, it only takes 2 lines of code. You can use ChatGPT to convert it into your own language.

<?php
$re = file_get_contents("php://input");
$json = json_decode($re);
print_r($json);
?>

Testing Method

1. In the "Source" Tab, select the relevant Action in the Input section.
2. Copy the Body Json for the corresponding feature from this document and paste it in.
3. Click "Start" to begin.

---

Security Notes

• Please be sure to set notifyHook; you will be notified automatically when image generation succeeds.
• getTask and listTask are only for querying if there are no generation results after 120 seconds.
• This program does not provide any guarantee regarding the copyright of the generated images.

---

If you have any questions, please contact: 51bobo [at] gmail.com