Stable Diffusion ForgeUI | 全新的SD版本，AI绘画神器，为拯救低显存用户而生 | 0到1安装教程！

AI工坊#十个骑士
15 Mar 202412:13

TLDR本次介绍的是一个名为Stable Diffusion WebUI Forge的新型绘画工具，它属于SD项目的一个分支。Forge项目在视频内存优化方面做了很多工作，对于显存较小的用户来说，使用Forge可以显著提升推理速度和绘图速度，同时减少GPU的使用率，并且扩散分辨率可以提高2-3倍。特别是对于6G显存的用户，提升速度可达60%到75%。安装过程简单，只需下载官方提供的一键安装包，解压后根据提示更新即可。此外，Forge还支持模型共享，可以避免重复下载，节省空间。在安装过程中，用户可以选择安装中文版本，以便更好地使用。Forge的特点在于其快速的绘图速度和优化的用户体验，适合显存较小的用户使用。

Takeaways

  • 🎨 **Stable Diffusion WebUI Forge** is a new painting tool that's part of the same project as the SD project, offering a branch with optimized video memory usage.
  • 🚀 **Performance Improvements**: Forge offers significant speed enhancements, with up to a 45% increase in inference speed and 2-3 times higher diffusion resolution for systems with 8GB of video memory.
  • 💻 **Optimized for Low Memory**: The tool is particularly beneficial for users with lower video memory, providing up to a 75% improvement in drawing speed for systems with 6GB of video memory.
  • 📈 **Minimal Gains for High Memory**: For systems with ample video memory, like 24GB, improvements are minimal, only offering a 3-6% increase.
  • 📦 **One-Click Installation**: Official one-click installation packages are available for ease of setup.
  • 🔄 **Update and Run**: After downloading, users should update to the latest version and then run the application.
  • 🌐 **Interface Language**: The interface defaults to English but can be switched to Chinese or other languages for user convenience.
  • 🔗 **Model Sharing**: Users can share models to avoid downloading them repeatedly, saving space and time.
  • 📁 **File Path Configuration**: Correct configuration of file paths is crucial for sharing models and ensuring the tool functions correctly.
  • 🔧 **Plugin Migration**: Existing plugins can be migrated to Forge, but caution is advised to avoid conflicts by ensuring only one instance of similar plugins is kept.
  • ⚙️ **Configuration Adjustments**: After installing language packs or migrating plugins, a restart of the UI interface is often required to apply changes.
  • 🖼️ **Fast Rendering**: The tool is noted for its fast rendering speeds, which is one of its standout features.

Q & A

  • What is the Stable Diffusion WebUI Forge?

    -The Stable Diffusion WebUI Forge is a new painting tool that belongs to the same project as the SD project. It is a branch project and shares the same layout and usage methods on the webui as the SD project.

  • What is the primary advantage of using the Forge tool?

    -The primary advantage of Forge is its speed and optimization on video memory. It increases inference speed by 30% to 45% for 8G video memory and can go up to 60% to 75% for 6G video memory.

  • How does Forge improve GPU usage and diffusion resolution?

    -Forge reduces the related usage of the GPU and increases the diffusion resolution by 2-3 times, which means it prevents memory overflow (OOM) and provides faster drawing speeds.

  • Is the Forge project suitable for users with large video memory?

    -The Forge project is more suitable for users with relatively small video memory. For users with large video memory, the speed improvement is minimal, and there is no significant need to use Forge.

  • What is the one-click installation package for Forge?

    -The one-click installation package for Forge is provided by the official source. It allows users to download and install the tool directly with ease.

  • How can users update the Forge tool after downloading it?

    -After downloading the Forge tool, users must first unzip the package and then follow the official prompts to update to the latest version by double-clicking the update button.

  • How can users change the interface language to Chinese?

    -To change the interface language to Chinese, users need to click the last item in the settings, then the third item, cancel the check mark, and click the button above to refresh all settings. After that, they can search for the Chinese version and install it.

  • What is the process to share a model in Forge?

    -To share a model in Forge, users need to open the Forge folder, find the appropriate file, edit it, and paste the path of their SD folder. After saving and restarting the UI, the model should be shared.

  • How can users migrate plugins from one tool to Forge?

    -Users can migrate plugins by copying all the plugins from the original tool's folder and pasting them into the same folder on the Forge side. After restarting the UI, the plugins should work normally.

  • What is the potential issue when migrating plugins?

    -When migrating plugins, there might be conflicts if there are plugins of the same type in both the original tool and Forge. It is recommended to keep only one instance of each plugin to avoid potential issues.

  • How does the Forge tool perform in terms of drawing speed and effect?

    -The Forge tool performs very well in terms of drawing speed and effect. It generates images quickly and with good quality, as demonstrated in the script.

Outlines

00:00

🖌️ Introduction to Stable Diffusion WebUI Forge

The video introduces the Stable Diffusion WebUI Forge, a new painting tool that is part of the same project as the SD project. The Forge is a branch project, sharing the same layout and usage methods as the SD's WebUI. The main advantage of Forge is its speed, with significant optimization in video memory leading to increased inference speed by 30% to 45% for 8GB of memory, and up to 60% to 75% for 6GB. It also reduces GPU usage and increases diffusion resolution by 2-3 times. The tool is particularly suitable for users with smaller video memory, as those with larger memory may not see substantial benefits from using Forge. The video also provides a step-by-step guide on how to download, update, and install the Forge, including changing the interface language to Chinese and troubleshooting installation issues.

05:00

🔄 Sharing and Setting Up Models in Forge

The second paragraph explains how to share and set up models within the Forge tool. It details the process of copying and pasting paths to share models, such as the SD folder, VAE, and ControlNet, and ensuring that the correct paths are used to avoid conflicts. The video demonstrates how to edit configuration files to reflect the shared model paths and emphasizes the importance of restarting the UI after making changes. It also shows how to troubleshoot and correct errors related to model sharing, such as ensuring that the correct file paths are used and that there are no leftover symbols from previous configurations. The paragraph concludes with a demonstration of the fast drawing speed and the successful sharing and use of different models in Forge.

10:01

📚 Migrate and Manage Plug-ins in Forge

The final paragraph focuses on migrating and managing plug-ins within the Forge environment. It outlines the process of copying installed plug-ins from one folder to another within the Forge directory, removing duplicates, and restarting the application to ensure that the plug-ins function correctly. The video also touches on potential issues that may arise from copying plug-ins, such as conflicts, and advises viewers to keep only one instance of similar plug-ins to avoid problems. The content concludes with a demonstration of creating an image using Forge, showcasing the tool's speed and effectiveness, and a reminder to viewers to try out the tool for themselves.

Mindmap

Keywords

💡Stable Diffusion WebUI Forge

Stable Diffusion WebUI Forge is a new painting tool that is part of the same project as the SD (Stable Diffusion) project. It is a branch project that shares the same layout and usage methods as the SD project's WebUI. The tool is optimized for video memory, making it faster and more efficient for users with lower video memory.

💡Optimization

Optimization in the context of the video refers to the improvements made to the tool to enhance its performance, particularly in terms of video memory usage. The Forge project has done significant optimization, which results in increased inference speed and drawing speed, and a decrease in GPU usage.

💡Inference Speed

Inference speed is the rate at which the AI tool processes and generates outputs based on input data. The video mentions that with 8GB of video memory, the inference speed can increase by 30% to 45%, making it a crucial aspect for users with limited video memory.

💡GPU Usage

GPU (Graphics Processing Unit) usage refers to the amount of processing power that the AI tool requires from the GPU to function. The Forge project reduces GPU usage, which is beneficial for users with lower-end graphics cards or limited GPU resources.

💡Diffusion Resolution

Diffusion resolution is the clarity or detail level at which the AI tool generates images. The video states that the Forge project can increase the diffusion resolution by 2-3 times, leading to higher quality image outputs.

💡OOM (Out of Memory)

OOM stands for 'Out of Memory,' which is a common issue in AI tools where the system runs out of available memory to process tasks. The Forge project addresses this by optimizing video memory usage, ensuring that the memory does not 'explode' or run out during operations.

💡One-Click Installation

One-click installation refers to a simplified installation process where users can set up the tool with a single action, making it accessible and user-friendly. The video mentions that the official provides a one-click installation package for the Forge project.

💡WebUI

WebUI stands for 'Web User Interface,' which is the graphical interface users interact with when using web-based applications. The video compares the WebUI of the Forge project to that of the SD project, noting their similarities.

💡Model Sharing

Model sharing is the process of making AI models accessible across different instances or installations of the tool. In the video, the presenter demonstrates how to share large models to avoid downloading them multiple times, which saves space and time.

💡Plug-in Migration

Plug-in migration involves transferring additional functionalities or extensions from one installation or version of the tool to another. The video discusses the migration of plug-ins from the SD project to the Forge project, emphasizing the need to avoid conflicts by managing duplicate plug-ins.

💡Chinese Version

The Chinese version refers to the localized interface of the tool that is presented in the Chinese language. The video script includes instructions on how to switch the interface language to Chinese, catering to a non-English speaking audience.

Highlights

Stable Diffusion WebUI Forge is introduced as a new painting tool.

Forge is a branch project of the SD project with similar layout and usage methods.

Optimization on video memory results in increased inference speed by 30% to 45% with 8GB memory.

Drawing speed with Forge is very fast, and GPU usage decreases.

Diffusion resolution can increase by 2-3 times with Forge.

Forge is particularly suitable for users with smaller video memory.

One-click installation package is available for easy setup.

After downloading, users must update to the latest version.

The interface can be switched to a Chinese version for ease of use.

The model 1.99G must be downloaded for the tool to function.

Sharing the model is possible to avoid re-downloading and save space.

Instructions on how to share the large model and set the path correctly are provided.

The VAE and ControlNet models can also be shared for convenience.

Migrating plugins from the original SD to Forge is straightforward.

A demonstration of creating an image with Forge showcases its fast drawing speed.

Users can change models within Forge for different drawing styles.

Migrating plugins requires caution to avoid conflicts.

The final output demonstrates the high quality and speed of Forge's image generation.