Source Diffusion 0.2.0-alpha.0

Whats changed

• More types of models can be loaded
• Changed the save format to save in metadata
• Changed the cache output by Diffusers to be saved in the application directory
• Changed the description when adding a model

Types of models that can be load

These are automatically identified, so there is no need to think about them when adding

• Stable Diffusion Model (Diffusers Model)
• Stable Diffusion Model (Safetensors Model)
• Stable Diffusion Model (CKPT Model)
• Stable Diffusion Vae (Diffusers Model)
• Stable Diffusion Vae (PT Model)
• Stable Diffusion Lora (Safetensors Model)
• Huggingface Repository Path

Note

The model loading process has changed significantly since this version, so if you have an earlier version installed, please delete the following files as they may cause bugs.

If you have already loaded models, you will find the model data in %APPDATA%\source-diffusion\models, so you will need to retrieve and load them again, unless you have loaded them from huggingface.

%APPDATA%\source-diffusion

Source Diffusion 0.1.5-alpha.0

Whats changed

Added ability to save all images.
For that, save the user data file.
Added ability to show seeds in history.

Source Diffusion 0.1.4-alpha.0

Whats changed

Optimized the processing of LoadWindow logs for improved performance.
Added support for running on Mac, however, image generation is not yet supported, so builds have not been performed.
Resolved the potential error caused by exceeding the buffer size for logs in the background process.

Source Diffusion 0.1.3-alpha.0

Whats changed

Fixed an issue where the application could not start because it couldn't find py_src, due to the current directory being set to System32 when launched directly from the installer.

Source Diffusion 0.1.2-alpha.0

What is changed

Added a button that allows you to reset Python when the application crashes on startup.
Added a check that ensures all necessary modules are installed before running the program. This will help to prevent errors caused by missing dependencies.

Source Diffusion 0.1.1-alpha.0

What is changed

This release introduces significant changes to the initial setup process and installation procedure.

Have addressed a longstanding issue with multiple instances running simultaneously. Our latest release ensures that the software can now only be run as a single instance, improving stability and preventing conflicts between different instances.

We hope these changes will improve your experience with our software. As always, please let us know if you encounter any issues or have any feedback.

Source Diffusion 0.1.0-alpha.0

Release 0.1.0-alpha.0

What is this

This software has been developed to make it easier for beginners to execute image generation that was previously difficult to start with a local environment in a web-based user interface for image generation. This software can be easily executed without the need for a command-line interface, however, please note that since this is an alpha version, there may be bugs or cases where it cannot be easily executed.

Support Platform

Windows 10, 11

Installation

1. Download the software.
2. Install the software using the MSI installer.

How to Use

1. Launch the software using the shortcut.
2. For the first launch, it may take some time due to the initial setup.
3. When the main window opens, import the model from the Models item on the right.
4. After importing, select a model from the list to load it.
5. Enter the prompt and generate the image!

For more detailed usage and support, please join our Discord server using the icon on the upper left.

Future Updates

We plan to fix bugs and add new features such as using Modals to allow the software to be run on PCs without a GPU, and developing a website for sharing generated images. We will provide support through Discord.