tenith/pokemon-lora

LoRA fine-tuned on custom dataset

Details

• Base model: stable-diffusion-v1-5/stable-diffusion-v1-5
• Resolution: 512px
• Method: LoRA (rank 16)
• Training steps: 1500
• Learning rate: 0.0001

Usage

import torch
from diffusers import StableDiffusionPipeline

pipe = StableDiffusionPipeline.from_pretrained(
    "stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch.float16
).to("cuda")

pipe.load_lora_weights("tenith/pokemon-lora")

image = pipe("your prompt here", num_inference_steps=30, guidance_scale=7.5).images[0]
image.save("output.png")