README.md

metadata

license: other
inference: true
tags:
  - stable-diffusion
  - stable-diffusion-diffusers
  - diffusers

This is a Custom Diffusion model fine-tuned from the Stable Diffusion v1-4.

Custom Diffusion allows you to fine-tune text-to-image diffusion models, such as Stable Diffusion, given a few images of a new concept (~4-20).

Here we give an example model fine-tuned using 5 images of a cat downloaded from UnSplash. The example code of inference is shown below.

Example code of inference

from diffusers import StableDiffusionPipeline
from src import diffuser_training

device = 'cuda'
model_id = "CompVis/stable-diffusion-v1-4"
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe = pipe.to(device)

diffuser_training.load_model(pipe.text_encoder, pipe.tokenizer, pipe.unet, 'custom_diffusion_cat.bin')

prompt = "<new1> cat swimming in a pool"
images = pipe(prompt, num_inference_steps=200, guidance_scale=6., eta=1.).images