Stable Diffusion v1-2

Recomendamos usar la biblioteca Diffusers de 🤗 para ejecutar Stable Diffusion.

pip install --upgrade diffusers transformers scipy

Running the pipeline with the default PNDM scheduler:
```python
import torch
from torch import autocast
from diffusers import StableDiffusionPipeline

model_id = "CompVis/stable-diffusion-v1-2"
device = "cuda"

pipe = StableDiffusionPipeline.from_pretrained(model_id)
pipe = pipe.to(device)

prompt = "a photo of an astronaut riding a horse on mars"
with autocast("cuda"):
    image = pipe(prompt)["sample"][0]  
    
image.save("astronaut_rides_horse.png")

Note:
Si estás limitado por la memoria de la GPU y tienes menos de 10GB de RAM disponible, asegúrate de cargar el StableDiffusionPipeline en precisión float16 en lugar de la precisión float32 predeterminada.
```python
import torch

pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe = pipe.to(device)

prompt = "a photo of an astronaut riding a horse on mars"
with autocast("cuda"):
    image = pipe(prompt, guidance_scale=7.5)["sample"][0]  
    
image.save("astronaut_rides_horse.png")

To swap out the noise scheduler, pass it to from_pretrained:

from diffusers import StableDiffusionPipeline, LMSDiscreteScheduler

model_id = "CompVis/stable-diffusion-v1-2"
# Use the K-LMS scheduler here instead
scheduler = LMSDiscreteScheduler(beta_start=0.00085, beta_end=0.012, beta_schedule="scaled_linear", num_train_timesteps=1000)
pipe = StableDiffusionPipeline.from_pretrained(model_id, scheduler=scheduler, use_auth_token=True)
pipe = pipe.to("cuda")

prompt = "a photo of an astronaut riding a horse on mars"
with autocast("cuda"):
    image = pipe(prompt, guidance_scale=7.5)["sample"][0]  
    
image.save("astronaut_rides_horse.png")