I recently got back into Ai Art, using Stable Diffusion. I was really blown away by the shear amount of options, and customization you can do within it's Ai engine. Stable Diffusion seems to really focus on portraits, so with that in mind, you have to become this master of prompts. I'm still learning, and have a long way to go, but you can begin to see the progression in prompts from my gens.(oldest to newest).
There are a few issues with Stable Diffusion, the biggest one is - it likes to generate nsfw, so carefully typing prompts is key to avoid all that, it also seems to focus heavily on females, while it can generate anything, males, nature, objects, females seem to be a huge part. Also Stable Diffusion doesnt handle distance very well, as well as hands. It tends to shine in close-up portraits. I tried my hardest to keep things wholesome and SFW. But everyone has a different idea of what's SFW, and NSFW, so there's your warning. (Hover to remove blur).