Generating an image

Now that we have the Stable Diffusion model loaded up to the GPU, let’s generate an image. text2img_pipe holds the pipeline object; all we need to provide is a prompt string, using natural language to describe the image we want to generate, as shown in the following code:

# generate an image
prompt ="high resolution, a photograph of an astronaut riding a horse"
image = text2img_pipe(
    prompt = prompt
).images[0]
image

Feel free to change the prompt to anything else that comes to your mind when you are reading this, for example, high resolution, a photograph of a cat running on the surface of Mars or 4k, high quality image of a cat driving a plane. It is amazing how Stable Diffusion can generate images according to a description in purely natural language.

If you run the preceding code without changing it, you may see an image like this showing up:

Figure 3.1: An image of an astronaut riding...