Nonnette

Introduction

Nonnette is a model in the Illustrious series, designed for text-to-image tasks using stable diffusion. It is created by merging and adjusting several existing models for personal use. The model is not intended to surpass existing ones but to provide a unique merge suitable for specific needs.

Architecture

Nonnette is a result of merging multiple models using the webui-reForge environment. The merging process involved various models, including lilybreed, kiwimixXL, catCarrier, and others, creating several versions until reaching the current Nonnette_v020. The model architecture follows the Illustrious series, utilizing specific prompts for quality and negative prompts to improve output.

Training

Nonnette has been fine-tuned using output images from other models and styles, such as lametta. The training involved several steps, combining styles and fine-tuning with specific datasets to achieve the desired artistic output. The process faced hardware limitations, impacting the training efficiency.

Guide: Running Locally

To run Nonnette locally, follow these steps:

1. Clone the Repository: Download the model files from the repository.
2. Set Up Environment: Ensure you have the necessary dependencies installed, typically involving Python and specific libraries for handling deep learning models.
3. Load the Model: Use a compatible framework like PyTorch to load the model.
4. Generate Images: Utilize text prompts to generate images, adjusting prompts for desired quality.

Cloud GPUs

For enhanced performance, consider using cloud GPU services such as AWS EC2, Google Cloud, or Azure, which provide powerful computing resources suitable for running large models.

License

Nonnette follows the Fair AI Public License 1.0 SD, inheriting from the Illustrious series' license and policies. For more details, refer to the licensing terms provided by Illustrious here.