Sign In
home
models
images
videos
posts
articles
bounties
tools
challenges
events
shop

ComfyUI Load Model With Previews !!! (Lora / Checkpoint / Lora Stack)

7
YoxRider's Avatar
YoxRider
Aug 22, 2024
workflows

English Version: Load Model with Previews

Introduction

This workflow is designed to streamline the process of importing AI models from local files and searching for corresponding previews on Civitai by calculating their BLAKE3 hash. By utilizing an integrated system of custom nodes, users can effortlessly load models from their local storage and automatically search for relevant previews on Civitai. This workflow is particularly useful for users who want to ensure they are working with the correct models and associated metadata, without the need for manual verification. With automatic preview generation and a straightforward setup, this workflow significantly enhances efficiency and user experience.

Required Nodes

  • was-node-suite-comfyui: Provides essential utilities and nodes for general operations.

  • ComfyUI_Comfyroll_CustomNodes: Adds custom functionalities tailored to specific tasks within ComfyUI.

  • ComfyUI-EasyCivitai-XTNodes: The core node suite that enables direct interaction with Civitai, including searching for models using BLAKE3 hash and importing previews.

  • ComfyUI-Easy-Use: Simplifies the overall workflow by providing user-friendly interfaces and tools for easier management and manipulation of models.

How to Use

  1. Setup: Ensure that all required nodes are installed (see installation instructions below).

  2. Configure: Open the workflow and input the desired model file path(s) in the appropriate node(s).

  3. Run: Execute the workflow. The system will automatically calculate the BLAKE3 hash of the local model file, search for the corresponding model on Civitai, generate previews, and integrate them into your project environment.

  4. Verify: Review the generated previews and loaded models to ensure accuracy and correctness.

This workflow is ideal for users who want a reliable method to integrate local models into their work while automatically searching for and confirming relevant previews from Civitai.

Installation Instructions for ComfyUI-EasyCivitai-XTNodes

Method A: Install via Civitai Manager

  1. Open Civitai Manager.

  2. Install via git url: https://github.com/X-T-E-R/ComfyUI-EasyCivitai-XTNodes

Method B: Manual Installation

  1. Clone the repository:

    git clone https://github.com/X-T-E-R/ComfyUI-EasyCivitai-XTNodes

  2. Install the required dependencies:

    pip install -r requirements.txt

  3. After installation, configure the system by modifying the following configuration files:

    • settings.toml: Contains general settings for the application.

    • .secrets.toml: Initially provided as .secrets.toml.example, needs to be copied and renamed to .secrets.toml.

    Note: An API token is required to download some NSFW previews from Civitai.

Screenshot

中文版本: 预览加载模型

介绍

此工作流旨在简化从本地文件导入AI模型，并通过计算BLAKE3哈希值在Civitai上查找相应预览的过程。通过整合一系列自定义节点，用户可以轻松地从本地存储加载模型，并自动搜索Civitai上相关的预览。这对于那些希望确保自己使用的是正确模型和相关元数据的用户特别有用，无需手动验证。借助自动预览生成和简化的设置，此工作流显著提高了工作效率和用户体验。

所需节点

  • was-node-suite-comfyui：提供基本的工具和节点，用于一般操作。

  • ComfyUI_Comfyroll_CustomNodes：添加了针对特定任务的自定义功能。

  • ComfyUI-EasyCivitai-XTNodes：核心节点套件，使得通过BLAKE3哈希值搜索模型和导入预览成为可能。

  • ComfyUI-Easy-Use：通过提供用户友好的界面和工具，简化整体工作流程，便于管理和操作模型。

如何使用

  1. 设置：确保所有所需节点均已安装（请参阅下文中的安装说明）。

  2. 配置：打开工作流，并在相应节点中输入所需的模型文件路径。

  3. 运行：执行工作流，系统将自动计算本地模型文件的BLAKE3哈希值，在Civitai上搜索相应的模型，生成预览，并将其集成到您的项目环境中。

  4. 验证：查看生成的预览和加载的模型，以确保其准确性和正确性。

此工作流非常适合那些希望在工作中集成本地模型，同时自动搜索并确认Civitai预览的用户。

ComfyUI-EasyCivitai-XTNodes安装说明

方法A：通过Civitai Manager安装

  1. 打开Civitai Manager。

  2. 直接通过git url安装：https://github.com/X-T-E-R/ComfyUI-EasyCivitai-XTNodes

方法B：手动安装

  1. 克隆代码库：

    git clone https://github.com/X-T-E-R/ComfyUI-EasyCivitai-XTNodes

  2. 安装所需依赖：

    pip install -r requirements.txt

  3. 安装完成后，通过修改以下配置文件来配置系统：

    • settings.toml：包含应用程序的一般设置。

    • .secrets.toml：最初作为.secrets.toml.example提供，需要复制并重命名为.secrets.toml。

    注意：某些NSFW预览的下载需要API令牌。

7

Comments