Flux Chill Guy Zone

Introduction

The Flux-Chill-Guy-Zone is a model designed for generating text-to-image outputs using diffusers and LoRA techniques. It features the character "Chill Guy" in various scenarios, offering a wide range of creative imagery.

Architecture

The model is based on the black-forest-labs/FLUX.1-dev architecture with specific parameters for image processing. It utilizes:

• Optimizer: AdamW
• Learning Rate Scheduler: Constant
• Noise Offset: 0.03
• Multires Noise Discount: 0.1
• Network Dimensions: 64
• Network Alpha: 32
• Repeat & Steps: 17 & 2600
• Epochs: 15
• Image Dimensions: Best at 768 x 1024, default 1024 x 1024

Training

The model is trained with 18 images utilizing the florence2-en label set. It is still in the training phase and may have artifacts or performance issues.

Guide: Running Locally

To run the model locally, follow these steps:

1. Setup Environment:

   • Ensure you have Python and PyTorch installed.
   • Install the required libraries using pip install transformers diffusers.

2. Import Libraries:

   import torch
   from pipelines import DiffusionPipeline
   

3. Load Model:

   base_model = "black-forest-labs/FLUX.1-dev"
   pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
   

4. Load Weights:

   lora_repo = "prithivMLmods/Flux-Chill-Guy-Zone"
   trigger_word = "chill guy"
   pipe.load_lora_weights(lora_repo)
   

5. Setup Device:

   device = torch.device("cuda")
   pipe.to(device)
   

6. Trigger Image Generation: Use the trigger word chill guy to generate images.

Cloud GPU

For faster performance, consider using cloud GPUs such as those available on AWS, Google Cloud, or Azure.

License

The model is licensed under the CreativeML OpenRAIL-M license, allowing for creative use with specific restrictions. Please refer to the license terms for more details before using the model.