My experiments with AI stuffses

Here's another experiment with AI image generation and some new feature testing.

Wallpaper generation with short prompts on SDXL

Stable Diffusion XL v1.0 Released with Additional Options on Clipdrop.co

Stable Diffusion XL (SDXL) is a newer, more powerful version of Stable Diffusion. It has a larger dataset and a more powerful algorithm, which results in sharper, more realistic, and more cohesive images. SDXL also has a number of new features, such as the ability to generate text within images and to create more descriptive images with shorter prompts.

The v1.0 release of SDXL is now available on the Clipdrop.co web client. This release includes a number of additional options, such as the ability to choose the aspect ratio of the generated images.

1. I decided to experiment with short prompts and wider aspect ratios to generate wallpaper-like images.

2. I wanted to see if I could create wide, panoramic images with short, concise prompts.

3. I was curious to see how the aspect ratio would affect the composition of the images.

4. I was also interested in seeing how the shorter prompts would affect the detail and realism of the images.

✨The prompt I used was "copenhagen cityscape, <xxxxx> style"

Here are the results!

Studio Ghibli

Graphic Novel

Pixar Movie

Futuristic Steampunk

Lego Bricks

Observations

1. The accuracy of short prompts is incredibly impressive.

2. The difference between using an image generator a few months ago and now is stark.

3. Making prompts easier to use is another step toward making this technology more accessible to everyone.

4. However, the accessibility of this technology also comes with the risks of generative AI.

5. Personally, I see myself using this to generate cover images for social media profiles.