Sign In

Using Google Gemini to create Prompts for Stable Diffusion

Using Google Gemini to create Prompts for Stable Diffusion

There are several extensions for SD that allow the use of chatgpt to create prompts for generating art. I wanted to see if the same could be accomplished with Google Gemini and if the results would be acceptable.

So I headed over to and I simply put "Create a prompt for generating art in stable diffusion" and hit enter. A second later, I had a result:

From there I copied the Subject, scene, style and details (literally highlighted those sections at one time and hit CTRL-C) and then put them in to my Automatic1111 prompt field and hit generate. The result was more than ok.

So general art worked ok, how about with people?

This time the prompt for Gemini was: create a prompt for generating an image of a middle aged woman in stable diffusion

and the result:

Gemini even provided some additional details to use and some additional prompts (like: A detailed portrait of a Latina woman in her 50s, with short, brown hair with streaks of grey, and kind eyes sparkling with laughter. She wears a colorful scarf tied around her neck and stands in front of a window overlooking a bustling marketplace. )

The result of the first prompt:

In all, I'm pretty happy with the results from the prompts that Gemini is coming up with and I am going to start using some of them in with my other sources and whims for generating art.

If you have any good prompts that were created by AI drop them in the comments below.