my hobby is exploring new things, in this blog i am going tell about how to generate image using google collab
just open google collab, then install some depantancies
!pip -q install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
!pip -q install diffusers transformers accelerate safetensorsthen use this code
import torch
from diffusers import AutoPipelineForText2Image
from PIL import Image
from IPython.display import displaydevice = "cuda" if torch.cuda.is_available() else "cpu"
dtype = torch.float16 if device == "cuda" else torch.float32
model_id = "stabilityai/sd-turbo" # small & quick
pipe = AutoPipelineForText2Image.from_pretrained(
modelid, torchdtype=dtype, use_safetensors=True
).to(device)
// small memory optimizations
pipe.enableattentionslicing()
pipe.enablevaetiling()
prompt = "give logo for pwntrends.com , cyberpunk theme, simple"
image = pipe(
prompt,
numinferencesteps=4, # 1–4 for SD-Turbo
guidance_scale=0.0, # 0.0–1.5 works best
width=512, height=512
).images[0]
image.save("/content/logo1.png")
display(image)
print("Saved to /content/logo1.png")
in this code customize your prompt
prompt = "give logo for pwntrends.com , cyberpunk theme, simple"
oh no its generating meaning less image
eventhogh i write blogs , beacuse i explore new things thats it
thank yo you for reading my blog, i will meet you in my next blog
Bye!!!!!!!!!!!
Leave a Reply