🌌 Orion AI — Text-to-Image Model

orion_ai is a fine-tuned Stable Diffusion model capable of generating high-quality images from text prompts. This version was fine-tuned using LoRA adapters.

🔧 Usage Example

from diffusers import StableDiffusionPipeline
import torch

# Load the model
pipe = StableDiffusionPipeline.from_pretrained(
    "Muyumba/orion_ai",
    torch_dtype=torch.float16,
).to("cuda")

# Generate an image
prompt = "A fantasy castle in the clouds"
image = pipe(prompt, num_inference_steps=50).images[0]
image.save("fantasy_castle.png")

📋 Model Details

• Base Model: Stable Diffusion
• Fine-tuning Method: LoRA (Low-Rank Adaptation)
• Training Data: Custom dataset
• Resolution: 512x512 (recommended)

🎨 Example Prompts

• "A majestic dragon flying over a medieval city"
• "Cyberpunk street scene with neon lights"
• "Beautiful landscape with mountains and lake at sunset"
• "Portrait of a wise old wizard with a long beard"

⚙️ Inference Parameters

• Steps: 20-50 (recommended: 30)
• Guidance Scale: 7.5-15 (recommended: 10)
• Sampler: DPM++ 2M Karras or Euler a

📝 License

This model is released under the MIT License. Feel free to use it for both commercial and non-commercial purposes.

🙏 Acknowledgments

Thanks to the Stable Diffusion community and Hugging Face for providing the tools and infrastructure.