F L U X.1 Fill dev G G U F

Introduction

FLUX.1-FILL-DEV-GGUF is a model developed by SporkySporkness, designed for non-commercial use under the flux-1-dev-non-commercial-license. It serves as a quantization of the original model from Black Forest Labs, adapted for compatibility with ComfyUI and stable-diffusion-webui-forge.

Architecture

The model utilizes GGUF quantized formats including Q4_0, Q5_0, Q8_0, and F16. These formats are optimized for different performance and precision requirements, making them suitable for various applications within compatible user interfaces.

Training

This model was created using a fork of the ComfyUI-GGUF project. The original model sourced from Black Forest Labs was adapted following the guidelines provided in the ComfyUI-GGUF repository.

Guide: Running Locally

To run the FLUX.1-FILL-DEV-GGUF model locally, follow these steps:

1. Clone the Repository: Start by cloning the ComfyUI-GGUF fork used for quantization.

   git clone https://github.com/mhnakif/ComfyUI-GGUF/
   

2. Install Dependencies: Ensure that all necessary libraries and dependencies are installed.
3. Download the Model: Obtain the model files from the Hugging Face model page.
4. Configure the Environment: Set up your environment to ensure compatibility with the quantization formats.
5. Run the Interface: Utilize either ComfyUI or stable-diffusion-webui-forge to deploy the model.

For improved performance, it is recommended to use cloud GPUs, such as those provided by AWS, Google Cloud, or Azure.

License

FLUX.1-FILL-DEV-GGUF is distributed under the flux-1-dev-non-commercial-license. For more details, refer to the license document.