---
library_name: diffusers
pipeline_tag: text-to-image
inference: true
base_model: stabilityai/stable-diffusion-2-1
---
# DPO LoRA Stable Diffusion v2-1
Model trained with LoRA implementation of Diffusion DPO Read more [here](https://github.com/huggingface/diffusers/tree/main/examples/research_projects/diffusion_dpo)
Base Model: https://huggingface.co/stabilityai/stable-diffusion-2-1
## Running with [🧨 diffusers library](https://github.com/huggingface/diffusers)
```python
from diffusers import DiffusionPipeline
from diffusers.utils import make_image_grid
import torch
pipe = DiffusionPipeline.from_pretrained(
"stabilityai/sd-turbo", # SD Turbo is a destilled version of Stable Diffusion 2.1
# "stabilityai/stable-diffusion-2-1", # for the original stable diffusion 2.1 model
torch_dtype=torch.float16, variant="fp16"
)
pipe.to("cuda")
pipe.load_lora_weights("radames/sd-21-DPO-LoRA", adapter_name="dpo-lora-sd21")
pipe.set_adapters(["dpo-lora-sd21"], adapter_weights=[1.0]) # you can play with adapter_weights to increase the effect of the LoRA model
seed = 123123
prompt = "portrait headshot professional of elon musk"
negative_prompt = "3d render, cartoon, drawing, art, low light"
generator = torch.Generator().manual_seed(seed)
images = pipe(
prompt=prompt,
negative_prompt=negative_prompt,
width=512,
height=512,
num_inference_steps=2,
generator=generator,
guidance_scale=1.0,
num_images_per_prompt=4
).images
make_image_grid(images, 1, 4)
```
## Guidance Scale vs LoRA weights
## Examples
Left Withoud DPO right with DPO LoRA
