Kabundji/sandie-lora

Sandie Fox LoRA

This is a LoRA (Low-Rank Adaptation) model trained for generating fox-related images, specifically featuring "Sandie" character/style.

Model Details

• Base Model: FLUX.1-dev
• Training Platform: Replicate
• Model Type: LoRA
• File Format: SafeTensors
• File Size: ~164MB

Usage

This model can be used with FLUX.1-dev base model for generating fox-themed images. Load the LoRA weights and use appropriate prompts to generate images featuring fox characteristics and the "Sandie" style.

Example Usage with Diffusers

from diffusers import FluxPipeline
import torch

# Load the base model
pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16)

# Load the LoRA weights
pipe.load_lora_weights("Kabundji/sandie-lora", weight_name="flux-lora.safetensors")
pipe.to("cuda")

# Generate image
prompt = "a beautiful fox in a forest, sandie style"
image = pipe(prompt, num_inference_steps=28, guidance_scale=3.5).images[0]
image.save("fox_output.png")

Training Details

• Trained using Replicate's infrastructure
• Optimized for fox character generation
• Compatible with FLUX.1-dev pipeline

License

Please check the licensing terms for the base model and ensure compliance with usage guidelines.

Credits

Model trained and uploaded by Kabundji.