Generating these images has been fun (though expensive!) and a real eye-opening experience into the nuance of prompt design and biases in artifical intelligence.
Over time I plan to add more DALL·E 3 images to allow comparisons with their equivelant DALL·E 2 versions.
“DALL·E 2 is a new AI system that can create realistic images and art from a description in natural language.”
That's a quote from the official website.
Essentially, you give it a text prompt (like “a cat in a hat”) and it will generate an image that attempts to match that prompt.
Click on an image to see the prompt that was used to generate it, along with the 3 additional images generated alongside it.
You can also click on the prompt text to navigate to a page that contains all 20 images (40 images if DALL·3 images exist for that prompt) generated for that prompt. It's interesting to see the styled shared across all images.
Feel free to follow me @dvyio on Threads.
|a representation of anxiety||an astronaut||the discovery of gravity||a horse||someone gazing at Mount Everest|