huacaya-alpaca

DreamBooth model for the huacaya concept trained by li-yan on the li-yan/alpaca-picture-dataset dataset.

This is a Stable Diffusion model fine-tuned on the huacaya concept with DreamBooth. It can be used by modifying the instance_prompt: a photo of huacaya alpaca on Golden Gate Bridge

This model was created as part of the DreamBooth Hackathon 🔥. Visit the organisation page for instructions on how to take part!

Description

This is a Stable Diffusion model fine-tuned on alpaca images for the animal theme.

Usage

%pip install -qqU diffusers accelerate

import torch
from diffusers import StableDiffusionPipeline

# Set device
device = (
    "mps"
    if torch.backends.mps.is_available()
    else "cuda"
    if torch.cuda.is_available()
    else "cpu"
)

# Load the pipeline
model_id = "li-yan/huacaya-alpaca"
pipe = StableDiffusionPipeline.from_pretrained(model_id).to(device)

# set prompt
prompt = "a photo of huacaya alpaca on the great wall" #@param

# Set up a generator for reproducibility
generator = torch.Generator(device=device).manual_seed(73)

# Run the pipeline, showing some of the available arguments
pipe_output = pipe(
    prompt=prompt, # What to generate
    negative_prompt="Oversaturated, blurry, low quality", # What NOT to generate
    height=480, width=640,     # Specify the image size
    guidance_scale=12,          # How strongly to follow the prompt
    num_inference_steps=50,    # How many steps to take
    generator=generator        # Fixed random seed
)

# View the resulting image
pipe_output.images[0]