Connect your ComfyUI to VidFlowLab Cloud Platform to unlock remote task scheduling capabilities. We provide two connection methods: Plugin Version (Recommended for normal users) and Worker Version (Recommended for servers/headless mode).
Best for users running ComfyUI on a local computer with a GUI.
Method A: Install via GitHub URL (ComfyUI Manager)
- Open ComfyUI Manager
- Click Install via Git URL
- Enter repo URL:
https://github.com/jacksnotes/VidFlowLab-ComfyUI-Bridge - Restart ComfyUI
Method B: Manual Installation (Recommended)
- Download VidFlowLab-ComfyUI-Bridge.zip from GitHub Releases.
- Unzip it into your
ComfyUI/custom_nodes/directory (Ensure the folder name isVidFlowLab-ComfyUI-Bridge). - Enter the folder and install dependencies:
pip install -r requirements.txt- Restart ComfyUI.
- Log in to VidFlowLab platform, go to "ComfyUI Bridge" page and generate an API Key.
- In ComfyUI canvas, Right Click -> Add Node ->
VidFlowLab->🔗 Bridge Settings. - Fill in the parameters:
- api_key: Your API Key (starts with
cb_) - server_url:
wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect
- api_key: Your API Key (starts with
- Check
save_settings, and trigger a run (Queue Prompt). - Seeing
Status: Connectedmeans connection is successful.
Best for headless servers (Linux) or long-running background tasks. It's a standalone Python script, not dependent on ComfyUI's plugin system.
Requires Python environment:
pip install websocket-client requestsEnsure your ComfyUI is already running (e.g., http://127.0.0.1:8188).
Run via Command Line
python bridge_worker.py \
--api-key cb_YOUR_API_KEY \
--server wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect \
--comfyui http://127.0.0.1:8188Run in Background (Linux/Mac)
Using nohup:
nohup python bridge_worker.py \
--api-key cb_YOUR_API_KEY \
--server wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect \
> bridge.log 2>&1 &Or use screen / tmux.
Create service file /etc/systemd/system/comfyui-bridge.service:
[Unit]
Description=ComfyUI Bridge Worker
After=network.target
[Service]
Type=simple
User=root
# Change to actual path
WorkingDirectory=/opt/comfyui-bridge-worker
# Change to your API Key
Environment=BRIDGE_API_KEY=cb_YOUR_API_KEY
Environment=BRIDGE_SERVER_URL=wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect
# Change to your ComfyUI URL
Environment=COMFYUI_URL=http://127.0.0.1:8188
ExecStart=/usr/bin/python3 bridge_worker.py
Restart=always
RestartSec=10
[Install]
WantedBy=multi-user.targetEnable Service:
sudo systemctl daemon-reload
sudo systemctl enable comfyui-bridge
sudo systemctl start comfyui-bridge
sudo systemctl status comfyui-bridgeQ: Connection status is always Disconnected? A:
- Check if API Key is copied correctly.
- Check if server_url matches exactly (
wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect). - If using Docker, ensure container has internet access.
Q: How to connect remote ComfyUI with Standalone Worker?
A: Use --comfyui http://REMOTE_IP:8188 argument. Ensure remote ComfyUI listens on 0.0.0.0 (--listen 0.0.0.0) and firewall ports are open.
将你的 ComfyUI 连接到 VidFlowLab 云平台,解锁远程任务调度能力。我们提供两种连接方式:插件版(推荐普通用户)和 Worker 独立版(推荐服务器/无头模式用户)。
适合在本地电脑运行 ComfyUI 的用户,直接集成在 ComfyUI 界面中。
方法 A:通过 GitHub URL 安装 (ComfyUI Manager)
- 打开 ComfyUI Manager
- 点击 Install via Git URL
- 输入仓库地址:
https://github.com/jacksnotes/VidFlowLab-ComfyUI-Bridge - 重启 ComfyUI
方法 B:手动安装 (推荐)
- 从 GitHub Releases 下载 VidFlowLab-ComfyUI-Bridge.zip
- 解压到
ComfyUI/custom_nodes/目录下(确保文件夹名为VidFlowLab-ComfyUI-Bridge) - 进入文件夹运行依赖安装:
pip install -r requirements.txt- 重启 ComfyUI。
- 登录 VidFlowLab 平台,在 "ComfyUI 桥接" 页面生成 API Key。
- 在 ComfyUI 画布中右键添加节点:
VidFlowLab->🔗 Bridge Settings。 - 填入参数:
- api_key: 你的 API Key (以
cb_开头) - server_url:
wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect
- api_key: 你的 API Key (以
- 勾选
save_settings,点击执行一次(Queue Prompt)。 - 看到
Status: Connected即表示连接成功。
适合无图形界面的服务器 (Linux) 或需要长期后台运行的场景。它是独立的 Python 脚本,不依赖 ComfyUI 插件系统,更轻量稳定。
需要 Python环境:
pip install websocket-client requests确保你的 ComfyUI 已经在运行(例如 http://127.0.0.1:8188)。
命令行直接运行
python bridge_worker.py \
--api-key cb_你的API密钥 \
--server wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect \
--comfyui http://127.0.0.1:8188后台运行 (Linux/Mac)
使用 nohup:
nohup python bridge_worker.py \
--api-key cb_你的API密钥 \
--server wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect \
> bridge.log 2>&1 &或者使用 screen / tmux。
创建服务文件 /etc/systemd/system/comfyui-bridge.service:
[Unit]
Description=ComfyUI Bridge Worker
After=network.target
[Service]
Type=simple
User=root
# 修改为实际代码路径
WorkingDirectory=/opt/comfyui-bridge-worker
# 修改为你的 API Key
Environment=BRIDGE_API_KEY=cb_你的密钥
Environment=BRIDGE_SERVER_URL=wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect
# 修改为你的 ComfyUI 地址
Environment=COMFYUI_URL=http://127.0.0.1:8188
ExecStart=/usr/bin/python3 bridge_worker.py
Restart=always
RestartSec=10
[Install]
WantedBy=multi-user.target启用服务:
sudo systemctl daemon-reload
sudo systemctl enable comfyui-bridge
sudo systemctl start comfyui-bridge
sudo systemctl status comfyui-bridgeQ: 连接一直是 Disconnected? A:
- 检查 API Key 是否正确复制。
- 检查 server_url 是否完全一致 (
wss://api.vidflowlab.com/api/v1/comfyui-bridge/ws/connect)。 - 如果是 Docker 环境,确保容器能访问互联网。
Q: 独立版如何连接远程 ComfyUI?
A: 使用 --comfyui http://远程IP:8188 参数,并确保远程 ComfyUI 启动时加了 --listen 0.0.0.0 参数且防火墙放行。