It seems like everyone is doing it now. With bing image generator being so good (although its limited by its moderation filters) and photoshop taking the talent out of morphing, I see a vast sea of AI driven content that pretty much all looks like the same tame stuff. I decided to go back to the roots of using Stable Diffusion, etc. to get results. If you have any suggestions (within whats allowed in the teen subcategory) let me know, otherwise ill just make and post what I like.

                                                                                                                                                               

Love seeing new faces....among other things.

I'd love to see some American brown-eyed brunettes. It seems to be very hard to get the AI to do this.

Also unzipped jean shorts are incredibly sexy.

I'm always curious to see how other creator tools use my prompts... so I'd request a set "carrying high". Of any theme. I try to keep things fresh and interesting... I think the carrying prompt has shown the most promise... but I too am looking for more varied pregnant belly. I got it once recently...posted it as the one off it is. OH! Also... use "confident" with a posture description... it might help avoid what I call the "globe hold" (which by the way is because when you google image search pregnancy...it's 98% of what you see. Just the 2 hands posing the globe.

(April 30, 2024, 4:54 pm)rogueangel007 Love seeing new faces....among other things.

I'd love to see some American brown-eyed brunettes. It seems to be very hard to get the AI to do this.

Also unzipped jean shorts are incredibly sexy.

I had to use my best judgement on what an "american brown-eyed brunette" looks like, but brown eyed brown haired girls came with ease. You're right that the unbuttoned shorts are a little challenging, but not impossible. The freedom of not having to dance around the content filters of BIG AI really helps you get what you really want from these images. 

                                                                                                   

Model / settings in sd?

Someone sent me a private message asking about some labor content, so I gave it a try. Here are some of the "safe to post" results. Sometimes I clean up the AIs work in photoshop. I retouched a few of these images, but most of them are the way the AI spit them out. 

                                                                                                                                           

Is anyone ever going to post their settings and models used? I don't even know how you guys get images this good on stable.

I have an anime style lora i been working on creating based on 0k0j0's art style for those that know who that is. Is a preg model with a more pronounced bellybutton for better outies. I been wanting to share some of the images i been making, but im not sure if that would be allowed here since its anime, or if that would be fine cuz of it being part of the teens look? Idk either way i made the 0k0j0 lora based on Pony for stable diffusion. worked well with the Forge WebUI. For those with 4gb graphics cards, i recommend checking it out for those who previously couldnt check it out.

More medical content is needed!

(May 4, 2024, 7:59 pm)bellyluv More medical content is needed!

+1. Bonus if you can include ultrasound images showing multiples.