>>6974741it's a modification of a prompt shared by another anon a while ago. this is on NAI.
(realistic:1.15), sketch, line-art, (masterpiece:1.2), (highest quality:1.1), (sharp focus:1.3), (sanpaku:1.1), cute, (young:0.8) (1girl:1.2), (petite:1),(from below shot:1.2), (looking down:1.2), upskirt, perspective, dynamic pose, nazi, waffen-ss, ss insignia, iron cross, ((black:1.2) ss uniform:1.2), (black garrison hat:1.1), black skirt, pantyhose (western comic (style)), (evil:1) smirk, (full face blush:0.8), (detailed face+eyes:1.1), (pink hair:1.2), (brown eyes:1.2), (bob cut hair:1.3), round breasts, (urban background:1.3), (detailed background:1.1), (by (Alphonse Mucha:0.5), by (Ilya Kuvshinov:0.9)), small eyes
Negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, (blurry:1.3), out of focus, censorship, ugly, old, (deformed:1.1), bra, (amateur drawing:1.2), odd, (fat:1.2), (wide hips:1), morphing, washed out, breast grab, groping, big breasts, (depth of field:1.2), (green:1), (green clothes:1.1), abstract, simple background, (parted lips:0.7)
Steps: 40, Sampler: DDIM, CFG scale: 12, Seed: 3007369675, Size: 960x1600, Denoising strength: 0.6, Clip skip: 2, First pass size: 448x704
For that batch I'm using DPM++ 2S a, toned down the initial (realistic:1.1) and detailed background demands a bit, fixed some of the specifics in both negative and positive to match Rebecca, and am running it on a dreambooth checkpoint made up of NAI trained on ~20 Rebecca for 2000 steps. I think the dreambooth training is probably causing a bit of the flat shaded simple background but otherwise should be copyable without it