mindlywork_ Acrylic World

Introduction

AcrylicWorld is a text-to-image model trained using the AI Toolkit by Ostris under the Glif Loradex program. It is designed to generate images based on textual prompts using the Acryl!ck trigger word.

Architecture

The model utilizes the black-forest-labs/FLUX.1-dev as its base model and incorporates diffusers and LoRA (Low-Rank Adaptation) techniques to enhance image generation capabilities.

Training

AcrylicWorld is fine-tuned from the FLUX.1-dev base model using specific prompts that include the trigger word Acryl!ck. This specialized training allows the model to produce diverse and creative outputs as demonstrated in various sample images.

Guide: Running Locally

To run the AcrylicWorld model locally:

1. Download the Model Weights: Access the model weights in Safetensors format from the Files & Versions tab.
2. Set Up Environment: Ensure you have Python and necessary libraries like PyTorch installed.
3. Load the Model: Use the Hugging Face Transformers library to load the model.
4. Provide Prompts: Use the Acryl!ck trigger word in your prompts for optimal image generation.

For enhanced performance, especially on large-scale tasks, consider using cloud-based GPUs such as those from Google Cloud or AWS.

License

AcrylicWorld is distributed under the flux-1-dev-non-commercial-license. Users must comply with the terms specified in the license, which allows non-commercial use.