【ComfyUI】python调用生图API，实现批量出图

官方给的示例：

https://github.com/comfyanonymous/ComfyUI/blob/master/script_examples/websockets_api_example.pyhttps://github.com/comfyanonymous/ComfyUI/blob/master/script_examples/websockets_api_example.pyhttps://github.com/comfyanonymous/ComfyUI/blob/master/script_examples/websockets_api_example.py

一、环境准备

1.1 comfyui环境安装参考

【本地运行AI绘画】ComfyUI的安装与使用（一）（windows+1660ti 6G显存）

1.2  调用可能需要安装的python包 

对应  import websocket

pip install websocket-client 
pip install json 

1.3 comfyui打开开发者模式，可以直接下载任意工作流 api文件

1.3.1 在英文最新界面打开

1.3.2  中文老界面打开

1.4 加载默认工作流，下载api，看看

保存结果是基本json 与工作流相似 不要弄混了

二、 代码

2.1  指定任意模型进行生图（找到名字和在api中的位置）

局部代码

def text_to_image(prompt_text,
                  model_name="SD1.X/dreamshaper_8.safetensors",
                  seed=6,
                  steps=20,
                  width=512,
                  height=256,
                  batch_size=2,
                  local_save_dir='./output',
                  api_file='api_demo.json'
                  
                  ):
    prompt = read_json(api_file)
    prompt["3"]["inputs"]["seed"] = seed  # 换一个数字不一样的图
    prompt["3"]["inputs"]["steps"] =steps   # 换一个数字不一样的图
    prompt["4"]["inputs"]["ckpt_name"] = model_name
    prompt["5"]["inputs"]["width"] = width
    prompt["5"]["inputs"]["height"] = height
    prompt["5"]["inputs"]["batch_size"]=batch_size

    prompt["6"]["inputs"]["text"] = prompt_text  

2.1.1  运行结果

2.1.2 参数解析

在comfyui的界面里确认有的模型名免得传错

打开保存的  api_demo.json,找到相应的代码改写

2.1.3 完整代码

#This is an example that uses the websockets api to know when a prompt execution is done
#Once the prompt execution is done it downloads the images using the /history endpoint

import websocket #NOTE: websocket-client (https://github.com/websocket-client/websocket-client) #  pip install websocket-client
import uuid
import json
import urllib.request
import urllib.parse
import os

server_address = "127.0.0.1:8188"
client_id = str(uuid.uuid4())

def queue_prompt(prompt):
    p = {"prompt": prompt, "client_id": client_id}
    data = json.dumps(p).encode('utf-8')
    req =  urllib.request.Request("http://{}/prompt".format(server_address), data=data)
    return json.loads(urllib.request.urlopen(req).read())

def get_image(filename, subfolder, folder_type):
    data = {"filename": filename, "subfolder": subfolder, "type": folder_type}
    url_values = urllib.parse.urlencode(data)
    with urllib.request.urlopen("http://{}/view?{}".format(server_address, url_values)) as response:
        return response.read()

def get_history(prompt_id):
    with urllib.request.urlopen("http://{}/history/{}".format(server_address, prompt_id)) as response:
        return json.loads(response.read())

def get_images(ws, prompt):
    prompt_id = queue_prompt(prompt)['prompt_id']
    output_images = {}
    while True:
        out = ws.recv()
        if isinstance(out, str):
            message = json.loads(out)
            if message['type'] == 'executing':
                data = message['data']
                if data['node'] is None and data['prompt_id'] == prompt_id:
                    break #Execution is done
        else:
            # If you want to be able to decode the binary stream for latent previews, here is how you can do it:
            # bytesIO = BytesIO(out[8:])
            # preview_image = Image.open(bytesIO) # This is your preview in PIL image format, store it in a global
            continue #previews are binary data

    history = get_history(prompt_id)[prompt_id]
    for node_id in history['outputs']:
        node_output = history['outputs'][node_id]
        images_output = []
        if 'images' in node_output:
            for image in node_output['images']:
                image_data = get_image(image['filename'], image['subfolder'], image['type'])
                images_output.append(image_data)
        output_images[node_id] = images_output

    return output_images

def read_json(api_file="controlnet_api.json"):
    # print('api_file',api_file)
    with open (api_file,"r",encoding="utf-8") as file_json:
        prompt_text=json.load(file_json)
    return prompt_text

def text_to_image(prompt_text,
                  model_name="SD1.X/dreamshaper_8.safetensors",
                  seed=6,
                  steps=20,
                  width=512,
                  height=256,
                  batch_size=2,
                  local_save_dir='./output',
                  api_file='api_demo.json'
                  
                  ):
    prompt = read_json(api_file)
    prompt["3"]["inputs"]["seed"] = seed  # 换一个数字不一样的图
    prompt["3"]["inputs"]["steps"] =steps   # 换一个数字不一样的图
    prompt["4"]["inputs"]["ckpt_name"] = model_name
    prompt["5"]["inputs"]["width"] = width
    prompt["5"]["inputs"]["height"] = height
    prompt["5"]["inputs"]["batch_size"]=batch_size

    prompt["6"]["inputs"]["text"] = prompt_text  

    ws = websocket.WebSocket()
    ws.connect("ws://{}/ws?clientId={}".format(server_address, client_id))
    images = get_images(ws, prompt)

    
    os.makedirs(local_save_dir, exist_ok=True)
    for node_id in images:
        for i,image_data in enumerate(images[node_id]):
            from PIL import Image
            import io
            image = Image.open(io.BytesIO(image_data))
            # image.show()
            save_path=f"{local_save_dir}/{prompt_text[:20]}_{i}.png"
            image.save(save_path)
            print(f"Saved image to {save_path}")

    
    ws.close()

if __name__ == "__main__":
    text_to_image(
                  prompt_text="a boy",
                  api_file=r'D:\deploy2\api_demo.json')