Prerequisites

Before starting, ensure you have the following:

• Google Account: You need a Google account to access Google Colab.
• Google Colab: Go to Google Colab and create a new notebook.
• GPU Runtime: Enable GPU for faster processing:
  • Go to Runtime > Change runtime type > Hardware accelerator > GPU.
• Stable Diffusion Model: We’ll use the runwayml/stable-diffusion-v1-5 model, which is pre-trained and available via the diffusers library.
• Python Libraries: Install the required libraries using pip (covered in Step 2).

Step 1: Open Google Colab

Go to Google Colab and create a new notebook.

Ensure you are using a GPU runtime for faster processing:

• Go to Runtime > Change runtime type > Hardware accelerator > GPU.

Step 2: Install Required Libraries

Install the necessary libraries using pip:

!pip install diffusers transformers accelerate torch

Step 3: Import Required Libraries

Import the libraries into your notebook:

import torch
from diffusers import StableDiffusionPipeline
from PIL import Image

Step 4: Load the Stable Diffusion Model

Load the Stable Diffusion model:

# Load the Stable Diffusion pipeline
pipe = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipe = pipe.to("cuda")

Step 5: Define Your Text Prompt

Define the text prompt for the image you want to generate:

prompt = "A futuristic cityscape at sunset with flying cars and neon lights, highly detailed, 4k resolution"

Step 6: Generate the Image

Generate the image using the prompt:

# Generate the image
with torch.autocast("cuda"):
    image = pipe(prompt).images[0]

Step 7: Display the Image

Display the generated image:

image.show()

Step 8: Save the Image

Save the image to your local machine or Google Drive:

Option 1: Save to Local Machine

image.save("generated_image.png")

Option 2: Save to Google Drive

from google.colab import drive
drive.mount('/content/drive')
image.save("/content/drive/My Drive/generated_image.png")

Step 9: Experiment with Different Prompts

Try different prompts to generate various types of images. For example:

• "A magical forest with glowing mushrooms and fairies"
• "A steampunk airship flying over a desert"
• "A realistic portrait of a lion with a majestic mane"

Full Code for Reference

# Step 1: Install libraries
!pip install diffusers transformers accelerate torch

# Step 2: Import libraries
import torch
from diffusers import StableDiffusionPipeline
from PIL import Image

# Step 3: Load the model
pipe = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
pipe = pipe.to("cuda")

# Step 4: Define the prompt
prompt = "A futuristic cityscape at sunset with flying cars and neon lights, highly detailed, 4k resolution"

# Step 5: Generate the image
with torch.autocast("cuda"):
    image = pipe(prompt).images[0]

# Step 6: Display the image
image.show()

# Step 7: Save the image (optional)
image.save("generated_image.png")

# Step 8: Save to Google Drive (optional)
from google.colab import drive
drive.mount('/content/drive')
image.save("/content/drive/My Drive/generated_image.png")

Tips for Better Results

• Be Descriptive: Use detailed prompts to guide the model.
• Adjust Image Size: Modify the image size using height and width parameters.
• Experiment with Styles: Add style keywords like “realistic,” “cartoon,” or “cyberpunk.”

Troubleshooting

• Out of Memory Error: Reduce the image size or use a simpler prompt.
• Slow Performance: Ensure you are using a GPU runtime in Colab.