Since I was asked for prompts, this is made using a LoRA I trained, you can find it here:
https://huggingface.co/Aotsuyu/loraDumpster/tree/main
I'm also posting the prompt for the first picture, using my LoRA you should be able to achieve different poses and outfits as well.

(extremely detailed CG unity 8k wallpaper), (best quality), (ultra-detailed), qiyana, smug, open mouth, (dark-skinned female), (dark skin:1.3)orange eyes, white hair, medium hair, green dress, short dress, strapless, tiara, armlet, jewelry, necklace, pantyhose, low neckline, gold bracelet, gem, cleavage wide hips, green eyeshadow, (makeup:0.8), blunt bangs, (thick thighs:0.7), green nails, (From Behind, Looking At Viewer:0.5), (Paw Pose, standing:0.5), perfect hands, simple background
Negative prompt: (low quality, worst quality:1.4), (bad anatomy), by bad-artist, bad-hands-5, bad-image-v2-39000, bad hands
Steps: 35, Sampler: DPM++ SDE Karras, CFG scale: 7, Seed: 2022892065, Size: 552x752, Model hash: f30926bf30, Denoising strength: 0.37, Clip skip: 2, ENSD: 31337,  AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: qiyana-epoch-000006(f2d25e93), AddNet Weight 1: 0.75, Hires upscale: 2.3, Hires steps: 15, Hires upscaler: 4x-AnimeSharp, Discard penultimate sigma: True

The LoRA is not as clean as my Ozen LoRA I feel, so that's why it's in the dumpster for now.