Redshift Diffusion

🚀 Redshift Diffusion

Redshift Diffusion is a fine - tuned Stable Diffusion model trained on high - resolution 3D artworks. It allows users to generate unique images by using the redshift style token in prompts.

✨ Features

• High - quality 3D Art Generation: Trained on high - resolution 3D artworks, it can generate stunning 3D - style images.
• Easy - to - use Token: Simply use the redshift style token in your prompts to achieve the desired effect.

📦 Installation

This model can be used just like any other Stable Diffusion model. For more information, please have a look at the Stable Diffusion. You can also export the model to ONNX, MPS and/or FLAX/JAX.

💻 Usage Examples

Basic Usage

from diffusers import StableDiffusionPipeline
import torch

model_id = "nitrosocke/redshift-diffusion"
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe = pipe.to("cuda")

prompt = "redshift style magical princess with golden hair"
image = pipe(prompt).images[0]

image.save("./magical_princess.png")

Advanced Usage

Prompt and settings for Tony Stark:

(redshift style) robert downey jr as ironman Negative prompt: glasses helmet Steps: 40, Sampler: DPM2 Karras, CFG scale: 7, Seed: 908018284, Size: 512x704

Prompt and settings for the Ford Mustang:

redshift style Ford Mustang Steps: 20, Sampler: DPM2 Karras, CFG scale: 7, Seed: 579593863, Size: 704x512

📚 Documentation

The name origin: I used Cinema4D for a very long time as my go - to modeling software and always liked the redshift render it came with. That is why I was very sad to see the bad results base SD has connected with its token. This is my attempt at fixing that and showing my passion for this render engine.

Characters rendered with the model: Cars and Landscapes rendered with the model:

If you enjoy my work and want to test new models before release, please consider supporting me

🔧 Technical Details

This model was trained using the diffusers based dreambooth training by ShivamShrirao using prior - preservation loss and the train - text - encoder flag in 11,000 steps.

📄 License

This model is open access and available to all, with a CreativeML OpenRAIL - M license further specifying rights and usage. The CreativeML OpenRAIL License specifies:

1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content
2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
3. You may re - distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL - M to all your users (please read the license entirely and carefully) Please read the full license here

Gradio

We support a Gradio Web UI run redshift - diffusion: