当地搭建自己的Midjourney API 申请及使用

当地搭建自己的Midjourney API 申请及使用

Midjourney 是一款非常强盛的 AI 绘图工具，只要输入关键字，就能在短短一两分钟生成十分精美的图像。Midjourney 以其出色的绘图能力在业界独树一帜，如今，Midjourney 早已在各个行业和领域广泛应用，其影响力愈发显著。
本文档重要先容 Midjourney API 中 Imagine 操作的使用流程，利用它我们可以轻松通过文本生成所需要的图像。
申请流程

要使用 Midjourney Imagine API，首先可以到 Midjourney Imagine API 页面点击「Acquire」按钮，获取请求所需要的凭证：

假如你尚未登录或注册，会自动跳转到登录页面邀请您来注册和登录，登录注册之后会自动返回当前页面。
在初次申请时会有免费额度赠送，可以免费使用该 API。
根本使用

接下来就可以在界面上填写对应的内容，如图所示：

在第一次使用该接口时，我们至少需要填写两个内容，一个是 authorization，直接在下拉列表内里选择即可。另一个参数是 prompt， prompt 就是我们想生成的图片描述内容，发起用英文描述，画的图会更正确结果更好，这里我们用了示例内容 Lamborghini speeds inside a volcano，代表要画一个兰博基尼在火山飞驰。
同时您可以注意到右侧有对应的调用代码生成，您可以复制代码直接运行，也可以直接点击「Try」按钮进行测试。

调用之后，我们发现返回结果如下：

1. {
2.   "image_url": "https://midjourney.cdn.acedata.cloud/attachments/1233387694839697411/1234197197067915365/36rgqit64j90qptsxnyq_Lamborghini_speeds_inside_a_volcano_id0494_f47263b6-ff92-44a3-88ee-51cf0e706aae.png?ex=662fdb36&is=662e89b6&hm=ca9be54907726937ed02517a13466bef2afb2825b7cda4b313de56a3c3310d0d&width=1024&height=1024",
3.   "image_width": 1024,
4.   "image_height": 1024,
5.   "image_id": "1234197197067915365",
6.   "raw_image_url": "https://midjourney.cdn.acedata.cloud/attachments/1233387694839697411/1234197197067915365/36rgqit64j90qptsxnyq_Lamborghini_speeds_inside_a_volcano_id0494_f47263b6-ff92-44a3-88ee-51cf0e706aae.png?ex=662fdb36&is=662e89b6&hm=ca9be54907726937ed02517a13466bef2afb2825b7cda4b313de56a3c3310d0d&",
7.   "raw_image_width": 2048,
8.   "raw_image_height": 2048,
9.   "progress": 100,
10.   "actions": [
11.     "upscale1",
12.     "upscale2",
13.     "upscale3",
14.     "upscale4",
15.     "reroll",
16.     "variation1",
17.     "variation2",
18.     "variation3",
19.     "variation4"
20.   ],
21.   "task_id": "1bae3bec-3ac4-4180-a148-74ee6cb68b98",
22.   "success": true
23. }

复制代码

返回结果一共有多个字段，先容如下：

• task_id，生成此图像任务的 ID，用于唯一标识此次图像生成任务。
  
• image_id，图片的唯一标识，在下次需要对图片进行变换操作时需要传此参数。
  
• image_url，缩略图的 URL，直接打开即可检察生成的结果。
  
• image_width：缩略图的像素宽度。
  
• image_height：缩略图的像素高度。
  
• raw_image_url：原图的 URL，和缩略图内容一样，但相比缩略图更加高清，加载速率会更慢一些。
  
• raw_image_width：原图的像素宽度。
  
• raw_image_height：原图的像素高度。
  
• actions，可以对生成的图片进行的进一步操作列表。这里一共列了 8 个，其中 upscale 代表放大，variation 代表变换。所以 upscale1 代表的就是对左上角第一张图片进行放大操作，variation3 就是代表根据左下角第三张图片进行变换操作。
  

打开 image_url 或者 raw_image_url 所对应的链接，可以发现如图所示。

可以看到，这里生成了一张 2x2 的预览图。到现在为止，第一次 API 调用就完成了。
图像放大与变换

下面我们实行针对当宿世成的照片进行进一步的操作，好比我们以为右上角第二张的图片还不错，但我们想进行一些变换微调，那么就可以进一步将 action 填写为 variation2，同时将 image_id 传递即可：

这时候得到的结果如下：

1. {
2.   "image_url": "https://midjourney.cdn.acedata.cloud/attachments/1233387694839697411/1234201336543969401/36rgqit64j90qptsxnyq_Lamborghini_speeds_inside_a_volcano_id0494_10dc56a7-ec16-4bac-878e-2338f2ae5f5d.png?ex=662fdf10&is=662e8d90&hm=9aec96bca35ae20b6f9ab536101b9c4ea255eb6216cbf7000ac554937da071f3&width=1024&height=1024",
3.   "image_width": 1024,
4.   "image_height": 1024,
5.   "image_id": "1234201336543969401",
6.   "raw_image_url": "https://midjourney.cdn.acedata.cloud/attachments/1233387694839697411/1234201336543969401/36rgqit64j90qptsxnyq_Lamborghini_speeds_inside_a_volcano_id0494_10dc56a7-ec16-4bac-878e-2338f2ae5f5d.png?ex=662fdf10&is=662e8d90&hm=9aec96bca35ae20b6f9ab536101b9c4ea255eb6216cbf7000ac554937da071f3&",
7.   "raw_image_width": 2048,
8.   "raw_image_height": 2048,
9.   "progress": 100,
10.   "actions": [
11.     "upscale1",
12.     "upscale2",
13.     "upscale3",
14.     "upscale4",
15.     "reroll",
16.     "variation1",
17.     "variation2",
18.     "variation3",
19.     "variation4"
20.   ],
21.   "task_id": "f4961620-1104-409f-9dc1-ba3ed15c2f4d",
22.   "success": true
23. }

复制代码

打开 image_url，新生成的图片如下所示：

可以看到，针对上一张右上角的图片，我们再次得到了四张类似的照片。
这时候我们可以挑选其中一张进行精致化地放大操作，好比选第四张，那就可以 action 传入 upscale4，通过 image_id 再次传入当前图像的 ID 即可。

   注意： upscale 操作相比 variation 来说，Midjourney 的耗时会更短一些。
  返回结果如下：

1. {
2.   "image_url": "https://midjourney.cdn.acedata.cloud/attachments/1233387694839697411/1234202545208033400/36rgqit64j90qptsxnyq_Lamborghini_speeds_inside_a_volcano_id0494_34edc3f5-2bd0-4f5b-a372-03270b02289b.png?ex=662fe031&is=662e8eb1&hm=f8006c4d33a03dfd027dffe4eb46ab0d113a4910aef07497f0b335c8998b7858&width=512&height=512",
3.   "image_width": 512,
4.   "image_height": 512,
5.   "image_id": "1234202545208033400",
6.   "raw_image_url": "https://midjourney.cdn.acedata.cloud/attachments/1233387694839697411/1234202545208033400/36rgqit64j90qptsxnyq_Lamborghini_speeds_inside_a_volcano_id0494_34edc3f5-2bd0-4f5b-a372-03270b02289b.png?ex=662fe031&is=662e8eb1&hm=f8006c4d33a03dfd027dffe4eb46ab0d113a4910aef07497f0b335c8998b7858&",
7.   "raw_image_width": 1024,
8.   "raw_image_height": 1024,
9.   "progress": 100,
10.   "actions": [
11.     "upscale_2x",
12.     "upscale_4x",
13.     "variation_subtle",
14.     "variation_strong",
15.     "zoom_out_2x",
16.     "zoom_out_1_5x",
17.     "pan_left",
18.     "pan_right",
19.     "pan_up",
20.     "pan_down"
21.   ],
22.   "task_id": "03f62b17-a6f1-4c8e-9b4d-1fc7bd5b1180",
23.   "success": true
24. }

复制代码

其中 image_url 如图所示：

如许我们就乐成得到了一张兰博基尼的照片。
同时注意到 actions 内里又包含了几个可进行的操作，先容如下：

• upscale_2x：对画面放大 2 倍，得到 2 倍高清图。
  
• upscale_4x：对画面放大 4 倍，得到 4 倍高清图。
  
• zoom_out_2x：对画面进行缩小两倍操作（周围区域填充）。
  
• zoom_out_1_5x：对画面进行缩小 1.5 倍操作（周围区域填充）。
  
• pan_left：对画面进行左偏移操作。
  
• pan_right：对画面进行右便宜操作。
  
• pan_up：对画面进行上偏移操作。
  
• pan_down：对画面进行下偏移操作。
  

可以继续按照上述流程传入对应的变换指令进行连续生图操作。
图像改写（垫图）

该 API 也支持图像改写，俗称垫图，我们可以输入一张图片 URL 以及需要改写的描述笔墨，该 API 就可以返回改写后的图片。
   注意：输入的图片 URL 需要是一张纯图片，不能是一个网页内里展示一张图片，否则无法进行图像改写。发起使用图床来上传获取图片的 URL。
  例如，我们这里有一张公路夕阳的图片，公路旁边有一些树木和楼房，如图所示：

现在我们想在它的底子上改写成海滩旁边，同时放一辆汽车停在路边。我们就可以构造如下的 prompt：

1. https://cdn.acedata.cloud/v014oc.png an illustration of a car parked on the beach --iw 2

复制代码

可以看到，我们的 prompt 的最开头是一个 HTTPS 开头的图片链接，然后接着加一个空格，后面跟上 prompt 笔墨的内容。这里我们还用了额外的一些高级参数，如 —iw 2 来调解图片的权重。
我们可以将如上内容作为一个整体，传递给 prompt 字段，如图所示：

输出结果如下：

1. {
2.   "image_url": "https://midjourney.cdn.acedata.cloud/attachments/1234427310434947145/1234539663515975690/atmateosa5693_An_illustration_of_a_car_parked_on_the_beach_id26_cc8650ec-7e4b-4685-8911-78172430d8a7.png?ex=66311a28&is=662fc8a8&hm=c39707a1f22bc7f12874060ea6ed58ba37c188139ccc9a13c61ed9f37e66ea74&width=1456&height=816",
3.   "image_width": 1456,
4.   "image_height": 816,
5.   "image_id": "1234539663515975690",
6.   "raw_image_url": "https://midjourney.cdn.acedata.cloud/attachments/1234427310434947145/1234539663515975690/atmateosa5693_An_illustration_of_a_car_parked_on_the_beach_id26_cc8650ec-7e4b-4685-8911-78172430d8a7.png?ex=66311a28&is=662fc8a8&hm=c39707a1f22bc7f12874060ea6ed58ba37c188139ccc9a13c61ed9f37e66ea74&",
7.   "raw_image_width": 2912,
8.   "raw_image_height": 1632,
9.   "progress": 100,
10.   "actions": [
11.     "upscale1",
12.     "upscale2",
13.     "upscale3",
14.     "upscale4",
15.     "reroll",
16.     "variation1",
17.     "variation2",
18.     "variation3",
19.     "variation4"
20.   ],
21.   "task_id": "24a79e8b-a79d-471a-aef7-089dc0627ee8",
22.   "success": true
23. }

复制代码

这时候我们就得到了如下生成的图片：

可以看到，在原来的图片整体风格和构图不变的前提下，整个场景变成了海滩旁边，同时公路上还出现了汽车，这就是 Prompt with Image。
图像融合

该 API 也支持图像融合，我们可以传入多张图片，以实现不同的图片融合结果。
好比说这里我们一共有两张图片，一张是一只玩具熊，另一张是一个电锯，分别如图所示：

现在我们想把二者融合起来，让这只熊拿着这个电锯，怎么做呢？
我们可以构造如下的 prompt：

1. https://i-blog.csdnimg.cn/img_convert/232d9019361dc53bd1578befcfa47a32.png https://i-blog.csdnimg.cn/img_convert/3d50c63b9c41701f46a9203a8e70bf39.png The bear is holding the chainsaw --iw 2

复制代码

可以发现，和 Image with Prompt 类似，我们这里将多张图片 URL 放在了 prompt 开头，并以空格分隔，末了再加上笔墨 prompt，将如上内容作为一个整体传递给 prompt 参数，运行结果如下：

1. {
2.   "image_url": "https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234547236830973972/kcisok_The_bear_is_holding_the_chainsaw_id8873344_ad605bc4-ba19-4807-b94f-367dab672f7a.png?ex=66312136&is=662fcfb6&hm=0fb1e2261c9a30b04de9da9b23b7562eb73677f1bbda1fae52c7243b12d25aac&width=1024&height=1024",
3.   "image_width": 1024,
4.   "image_height": 1024,
5.   "image_id": "1234547236830973972",
6.   "raw_image_url": "https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234547236830973972/kcisok_The_bear_is_holding_the_chainsaw_id8873344_ad605bc4-ba19-4807-b94f-367dab672f7a.png?ex=66312136&is=662fcfb6&hm=0fb1e2261c9a30b04de9da9b23b7562eb73677f1bbda1fae52c7243b12d25aac&",
7.   "raw_image_width": 2048,
8.   "raw_image_height": 2048,
9.   "progress": 100,
10.   "actions": [
11.     "upscale1",
12.     "upscale2",
13.     "upscale3",
14.     "upscale4",
15.     "reroll",
16.     "variation1",
17.     "variation2",
18.     "variation3",
19.     "variation4"
20.   ],
21.   "task_id": "891f2645-ee15-4c7b-ac24-d98163c8e57e",
22.   "success": true
23. }

复制代码

我们就得到了如下结果：

可以看到，我们就乐成实现了图片融合。
   注意：图片融合最多可以支持 5 个图片 URL 作为输入，也就是最多支持 5 张图片融合，输入格式同上。
  局部变换

该 API 也支持图像的局部绘图功能，但只支持在上文内容下生成的图片，我们可以传入一张生成图片的的唯一ID，局部重绘的行为参数 action 以及需要重绘区域的掩码mask，以实现在改掩码区域进行重绘。
好比说这里我们有一张生成关于猫的图片：

现在我们想对这个猫脸进行重绘，怎么做呢？
首先我们需要获取该区域的掩码mask，此mask是通过灰度图片进行Base64编码得来的，下面提供一些获取掩码的工具代码：
Python获取掩码示例代码：

1. import sys
2. import os
3. from PySide6.QtWidgets import *
4. from PySide6.QtGui import QPainter, QMouseEvent, QPen, QColor, QImage
5. from PySide6.QtCore import Qt, QPoint
7. class DrawingWidget(QWidget):
8.     def __init__(self, imagePath):
9.         super().__init__()
10.         self.setAttribute(Qt.WA_StaticContents)
11.         self.background_image = QImage(imagePath)
12.         imageSize = self.background_image.size()*0.8
13.         self.setFixedSize(imageSize)
14.         self.foreground_image = QImage(self.size(), QImage.Format_ARGB32)
15.         self.foreground_image.fill(Qt.transparent)
16.         self.drawing = False
17.         self.lastPoint = QPoint()
18.         self.pen_color = QColor(255, 255, 255, 255)
19.         self.pen_size = 50
21.     def set_pen_size(self, size):
22.         self.pen_size = size
24.     def mousePressEvent(self, event: QMouseEvent):
25.         if event.button() == Qt.LeftButton:
26.             self.drawing = True
27.             self.lastPoint = event.pos()
29.     def mouseMoveEvent(self, event: QMouseEvent):
30.         if event.buttons() & Qt.LeftButton and self.drawing:
31.             painter = QPainter(self.foreground_image)
32.             painter.setRenderHint(QPainter.Antialiasing, True)
33.             pen = QPen(self.pen_color, self.pen_size,
34.                        Qt.SolidLine, Qt.RoundCap, Qt.RoundJoin)
35.             painter.setPen(pen)
36.             painter.drawLine(self.lastPoint, event.pos())
37.             self.lastPoint = event.pos()
38.             self.update()
40.     def mouseReleaseEvent(self, event: QMouseEvent):
41.         if event.button() == Qt.LeftButton and self.drawing:
42.             self.drawing = False
44.     def paintEvent(self, event):
45.         canvasPainter = QPainter(self)
46.         canvasPainter.drawImage(
47.             self.rect(), self.background_image, self.background_image.rect())
48.         canvasPainter.drawImage(
49.             self.rect(), self.foreground_image, self.foreground_image.rect())
51.     def save_image(self, path):
52.         self.foreground_image.save(path)
54. class MainWindow(QDialog):
55.     def __init__(self, imagePath):
56.         super().__init__()
57.         self.setWindowTitle("mask tool")
58.         self.drawing_widget = DrawingWidget(imagePath)
59.         self.currentPath = os.getcwd().replace("\", "/")
60.         self.tempPath = self.currentPath + "/temp.jpg"
61.         self.projectPath = ""
62.         self.setDone = False
64.     def init_ui(self):
65.         layout = QVBoxLayout()
66.         layout.addWidget(self.drawing_widget)
67.         controls_layout = QHBoxLayout()
68.         size_label = QLabel("pen size:")
69.         controls_layout.addWidget(size_label)
70.         self.size_slider = QSlider(Qt.Horizontal)
71.         self.size_slider.setMinimum(100)
72.         self.size_slider.setMaximum(400)
73.         self.size_slider.setValue(400)
74.         self.size_slider.valueChanged.connect(self.update_pen_size)
75.         controls_layout.addWidget(self.size_slider)
76.         self.lineEdit_addPromp = QLineEdit()
77.         layout.addWidget(self.lineEdit_addPromp)
78.         save_button = QPushButton("   Start partial redrawing   ")
79.         save_button.clicked.connect(self.save_image)
80.         controls_layout.addWidget(save_button)
81.         dont_button = QPushButton("   Cancel partial redraw   ")
82.         dont_button.clicked.connect(self.dont_image)
83.         controls_layout.addWidget(dont_button)
84.         layout.addLayout(controls_layout)
85.         self.setLayout(layout)
87.     def update_pen_size(self):
88.         size = self.size_slider.value()
89.         self.drawing_widget.set_pen_size(size)
91.     def save_image(self):
92.         tempImage = self.currentPath + "/temp.jpg"
93.         self.drawing_widget.save_image(self.tempPath)
94.         self.prompt = self.lineEdit_addPromp.text()
95.         self.setDone = True
96.         self.close()
98.     def dont_image(self):
99.         self.setDone = False
100.         self.close()
102.     def closeEvent(self, event):
103.         if self.setDone:
104.             pass
105.         else:
106.             self.setDone = False
107.         event.accept()
110. if __name__ == '__main__':
111.     imagePath = "test.png"
112.     app = QApplication(sys.argv)
113.     mainWindow = MainWindow(imagePath)
114.     mainWindow.init_ui()
115.     mainWindow.exec()

复制代码

通过以上代码可以得到掩码图片，在这个过程中我们需要注意掩码图片必须与原图片一样的尺寸，然后掩码图片中白色的区域就是需要重绘的区域，下面展示猫图片需要重绘的掩码图片与原图的对比：
原图片：

掩码图片：

末了我们还需要将掩码图片转换为基于Base64的编码，接下来提供转换Base64编码的代码：
Python转换Base64示例代码：

1. import cv2
2. import base64
4. image_path = 'temp.jpg'
5. gray_image = cv2.imread(image_path)
6. _, buffer = cv2.imencode('.jpg', gray_image)
7. base64_encoded = base64.b64encode(buffer).decode('utf-8')
8. with open('grayscale_image_base64.txt', 'w') as f:
9.     f.write(base64_encoded)
11. print("success!")

复制代码

  注意：上述 Python 代码描述了 mask 的生成过程，假如您要将其对接到您的产品中，请根据其原理编写对应语言的代码。
  通过以上代码，我们得到了需要重绘的掩码 mask ，下面我们还需要需要将参数 action 设置为 variation_region ，生成图片的ID image_id （获取该参数参考上文的内容），以及传入对应的掩码 mask ，其它参数信息如下：

• action：对图片进行操作的行为，此处为 variation_region ，表现对图片进行局部重绘。
  
• prompt：对图像进行局部重绘的描述词（可选参数）。
  
• image_id：图片的唯一标识，方便对图像进行局部重绘。
  
• mask：图片对应的掩码区域的base64编码（图片是上面image_id指定的）。
  

因此根据以上规则我们需要设置正确的的参数，参数 prompt 是一个非必填参数，这里为了方便对比将掩码区域的 prompt 设置为 A cute cat ，详细的参数设置如下图所示：

代码示例

可以发现，在页面右侧已经自动生成了各种语言的代码，如图所示：

部门代码示例如下：
CURL

1. curl -X POST 'https://api.acedata.cloud/midjourney/imagine' \
2. -H 'accept: application/json' \
3. -H 'authorization: Bearer {token}' \
4. -H 'content-type: application/json' \
5. -d '{
6.   "prompt": "A  cute cat ",
7.   "action": "variation_region",
8.   "image_id": "1265875488702726144",
9.   "mask": "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"
10. }'

复制代码

Python

1. import requests
3. url = "https://api.acedata.cloud/midjourney/imagine"
5. headers = {
6.     "accept": "application/json",
7.     "authorization": "Bearer {token}",
8.     "content-type": "application/json"
9. }
11. payload = {
12.     "prompt": "A  cute cat ",
13.     "action": "variation_region",
14.     "image_id": "1265875488702726144",
15.     "mask": "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"
16. }
18. response = requests.post(url, json=payload, headers=headers)
19. print(response.text)

复制代码

响应示例

请求乐成后，API 将返回换脸后端图片结果信息。例如：

1. {
2.   "image_url": "https://platform.cdn.acedata.cloud/midjourney/6c9450d8-1c22-4f85-a527-e7a7bfb4a61b.png?imageMogr2/thumbnail/!50p",
3.   "image_width": 1024,
4.   "image_height": 1024,
5.   "actions": [
6.     "upscale1",
7.     "upscale2",
8.     "upscale3",
9.     "upscale4",
10.     "reroll",
11.     "variation1",
12.     "variation2",
13.     "variation3",
14.     "variation4"
15.   ],
16.   "raw_image_url": "https://platform.cdn.acedata.cloud/midjourney/6c9450d8-1c22-4f85-a527-e7a7bfb4a61b.png",
17.   "raw_image_width": 2048,
18.   "raw_image_height": 2048,
19.   "progress": 100,
20.   "image_id": "1265876571323891712",
21.   "task_id": "6c9450d8-1c22-4f85-a527-e7a7bfb4a61b",
22.   "success": true
23. }

复制代码

可以发现这是对掩码区域的图像进行了重绘操作，返回的结果与上文的是一致的，结果如下图所示：

可以看到，我们就乐成实现了对生成的图片的自界说区域进行局部重绘。
异步回调

由于 Midjourney 生成图片需要期待一段时间，所以本 API 也默认设计为了长期待模式。但在部门场景下，长期待可能会带来一些额外的资源开销，因此本 API 也提供了异步 Webhook 回调的方式，当图片生成乐成或失败时，其结果都会通过 HTTP 请求的方式发送到指定的 Webhook 回调 URL。回调 URL 吸收到结果之后可以进行进一步的处理。
下面演示详细的调用流程。
首先，Webhook 回调是一个可以吸收 HTTP 请求的服务，开发者应该更换为自己搭建的 HTTP 服务器的 URL。此处为了方便演示，使用一个公开的 Webhook 样例网站 https://webhook.site/，打开该网站即可得到一个 Webhook URL，如图所示：

将此 URL 复制下来，就可以作为 Webhook 来使用，此处的样例为 https://webhook.site/995d0a91-d737-40a7-a3b9-5baf68ed924c。
接下来，我们可以设置字段 callback_url 为上述 Webhook URL，同时填入 prompt，如图所示：

点击测试之后会立即得到一个 task_id 的响应，用于标识当宿世成任务的 ID，如图所示：

稍等半晌，等图片生成竣事，可以发发现 Webhook URL 收到了一个 HTTP 请求，如图所示：

其结果就是当前任务的结果，内容如下：

1. {
2.   "success": true,
3.   "task_id": "f6e39eaf-652a-4bf5-a15c-79d8b143b80a",
4.   "image_url": "https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234551030549839932/kcisok_A_cat_sitting_on_a_table_id2724480_591c5c85-ec80-42ab-9fe5-9adfbed192e4.png?ex=663124be&is=662fd33e&hm=da725eb74aae375d60beec38b4cd26c5a7b373b1662f222ff838a8ea6fd5e798&width=1024&height=1024",
5.   "image_width": 1024,
6.   "image_height": 1024,
7.   "image_id": "1234551030549839932",
8.   "raw_image_url": "https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234551030549839932/kcisok_A_cat_sitting_on_a_table_id2724480_591c5c85-ec80-42ab-9fe5-9adfbed192e4.png?ex=663124be&is=662fd33e&hm=da725eb74aae375d60beec38b4cd26c5a7b373b1662f222ff838a8ea6fd5e798&",
9.   "raw_image_width": 2048,
10.   "raw_image_height": 2048,
11.   "progress": 100,
12.   "actions": [
13.     "upscale1",
14.     "upscale2",
15.     "upscale3",
16.     "upscale4",
17.     "reroll",
18.     "variation1",
19.     "variation2",
20.     "variation3",
21.     "variation4"
22.   ]
23. }

复制代码

其中 success 字段标识了该任务是否执行乐成，假如执行乐成，还会有同样的 actions, image_id, image_url 字段，和上文先容的返回结果是一样的，别的另有 task_id 用于标识任务，以实现 Webhook 结果和最初 API 请求的关联。
假如图片生成失败，Webhook URL 则会收到类似如下内容：

1. {
2.   "success": false,
3.   "task_id": "7ba0feaf-d20b-4c22-a35a-31ec30fc7715",
4.   "error": {
5.     "code": "bad_request",
6.     "message": "Unrecognized argument(s): `-c`, `x`"
7.   }
8. }

复制代码

这里的 success 字段会是 false，同时还会有 error.code 和 error.message 字段描述了任务错误的详情信息，Webhook 服务器根据对应的结果进行处理即可。
流式输出

Midjourney 官方在生成图片的时候是有进度的，在最开始是一张含糊的照片，然后经过几次迭代之后，图片徐徐变得清晰，末了得到完整的图片。
所以，一张图片的生成过程大约可以分为「发送死令」->「开始生图（多次迭代徐徐清晰）」->「生图完毕」的阶段。
在没开启流式输出的情况下，本 API 从发起请求到返回结果，实际上是从上述「发送死令」->「生图完毕」的全过程，中间生图的过程也全被包含在内里，由于 Midjourney 本身生成图片速率较慢，整个过程大约需要期待一分钟或更久。
所以为了更好的用户体验，本 API 支持流式输出，即当「开始生图」的时候就开始返回结果，每当绘制进度有变革，就会流式将结果输出，直至生图竣事。
假如想流式返反响应，可以更改请求头内里的 accept 参数，修改为 application/x-ndjson，不过调用代码需要有对应的更改才能支持流式响应。
Python 样例代码：

1. import requests
3. url = 'https://api.acedata.cloud/midjourney/imagine'
4. headers = {
5.     'content-type': 'application/json',
6.     'accept': 'application/x-ndjson',
7.     'authorization': 'Bearer {token}'
8. }
9. body = {
10.     "prompt": "a beautiful cat --v 6"
11. }
12. r = requests.post(url, headers=headers, json=body, stream=True)
13. for line in r.iter_lines():
14.     print(line.decode())

复制代码

运行结果：

1. {"image_url":"https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234558451443699803/eae94f0f-0ba5-4b3c-9bad-59fb33ac2cbc_grid_0.webp?ex=66312ba7&is=662fda27&hm=4625d5f12158bffc07c4faaf6ce75af6f1396122f148b33b91f3e20b48fecc8b&width=256&height=256","image_width":256,"image_height":256,"image_id":"1234558451443699803","raw_image_url":"https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234558451443699803/eae94f0f-0ba5-4b3c-9bad-59fb33ac2cbc_grid_0.webp?ex=66312ba7&is=662fda27&hm=4625d5f12158bffc07c4faaf6ce75af6f1396122f148b33b91f3e20b48fecc8b&","raw_image_width":512,"raw_image_height":512,"progress":35,"actions":[],"task_id":"49589d2c-b6b3-4fbf-9f82-93068509c76f","success":true}
2. {"image_url":"https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234558458595115149/eae94f0f-0ba5-4b3c-9bad-59fb33ac2cbc_grid_0.webp?ex=66312ba9&is=662fda29&hm=9af53fa645127131a88dfbb3930add7abda710c12a3d6c30c457d6a067b36bab&width=256&height=256","image_width":256,"image_height":256,"image_id":"1234558458595115149","raw_image_url":"https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234558458595115149/eae94f0f-0ba5-4b3c-9bad-59fb33ac2cbc_grid_0.webp?ex=66312ba9&is=662fda29&hm=9af53fa645127131a88dfbb3930add7abda710c12a3d6c30c457d6a067b36bab&","raw_image_width":512,"raw_image_height":512,"progress":75,"actions":[],"task_id":"49589d2c-b6b3-4fbf-9f82-93068509c76f","success":true}
3. {"image_url":"https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234558483408490566/kcisok_A_landscape_painting_of_a_beautiful_sunset_id5963392_eae94f0f-0ba5-4b3c-9bad-59fb33ac2cbc.png?ex=66312baf&is=662fda2f&hm=185ea8f130806bf8bd96911bd251808455fd65596edcdb459f9b3cfd7860387c&width=1024&height=1024","image_width":1024,"image_height":1024,"image_id":"1234558483408490566","raw_image_url":"https://midjourney.cdn.acedata.cloud/attachments/1234291876639674388/1234558483408490566/kcisok_A_landscape_painting_of_a_beautiful_sunset_id5963392_eae94f0f-0ba5-4b3c-9bad-59fb33ac2cbc.png?ex=66312baf&is=662fda2f&hm=185ea8f130806bf8bd96911bd251808455fd65596edcdb459f9b3cfd7860387c&","raw_image_width":2048,"raw_image_height":2048,"progress":100,"actions":["upscale1","upscale2","upscale3","upscale4","reroll","variation1","variation2","variation3","variation4"],"task_id":"49589d2c-b6b3-4fbf-9f82-93068509c76f","success":true}

复制代码

可以看到，启用流式输出之后，返回结果就是逐行的 JSON 了。
在 Node.js 情况中，实现代码可写为如下：

1. const axios = require("axios");
3. const url = "https://api.acedata.cloud/midjourney/imagine";
4. const headers = {
5.   "content-type": "application/json",
6.   accept: "application/x-ndjson",
7.   authorization: "Bearer {token}",
8. };
9. const body = {
10.   prompt: "a beautiful cat --v 6",
11.   action: "generate",
12. };
14. axios
15.   .post(url, body, { headers: headers, responseType: "stream" })
16.   .then((response) => {
17.     console.log(response.status);
18.     response.data.on("data", (chunk) => {
19.       console.log(chunk.toString());
20.     });
21.   })
22.   .catch((error) => {
23.     console.error(error);
24.   });

复制代码

这些示例运行的结果都是相似的。
请注意，流式输出结果中有一个称为 progress 的字段，表现生成进度，范围从 0 到 100。假如需要，您可以在页面上显示这些信息。
   注意：当生成未完全完成时，actions 字段为空，表现无法对中间图像执行进一步处理操作。生成完成后，在生成过程中生成的 image_url 将被销毁。
  别的，您可以通过指定 accept=application/x-ndjson 的请求头和 callback_url 的请求体，将流式结果与异步回调联合起来，然后 callback_url 可以吸收到多个流式结果的 POST 请求。

免责声明：如果侵犯了您的权益，请联系站长，我们会及时删除侵权内容，谢谢合作！更多信息从访问主页：qidao123.com:ToB企服之家，中国第一个企服评测及商务社交产业平台。