C666D Pear Booty Pawg Krea2 v1 https://civitai.red/models/2909674/c666d-pear-booty-pawg-krea2-v1 This model is trained with a dataset consisting of 94 images of mixed resolutions. Some of the images are AI, and others photos. It creates female characters with a exagerrated lower body, and long legs. The poses in the dataset are mainly: standing, sitting, squatting, bent over, from behind, front, side, etc.etc. without full nudity, explicit nsfw / R18-rated content. But many images do contain high waist lingerie, tight dresses, yoga pants, skirt and shorts, croptops and other cute and sexy stuff. So a wide variety of clothing items. -- Trained triggers: 🔥🔥🔥 1. pear booty pawg - - Main trigger to create the hourglass shape with enlarged buttocks. 2. thick thighs - - creates wide thighs below the butt cheeks. 3. wide hips - - makes the butt cheeks and the hip bones wider, compared to the waist. Other triggers/tags: single alone lady / woman / girl transparent clothes yoga pants leggings cute makeup lips eyelashes amateur / professional photography hard / soft / warm lighting makeup eyeliner lipstick caucasian / turkish / african / brazil / european -- Testing and Usage: 🤖🤖🤖 1. I have tested the lora to a peak strength of +2, and haven't had any distortion or deformations yet using turbo models. So it can be used as a slider. Feel free to experiment. 2. the lora works best when the triggers are used in a array, seperated by comma's: e.g: "pear booty pawg, thick thighs, wide hips". 3. To get a long dress that covers the legs ommit/remove the words ass and thighs. Instead prompt: "lowerbody". The model will still keep the pear-shape body with the clothes on. 4. The images used in the dataset are in a modern style, but captioned manually to include a wide variety of objects, locations, and other elements, so it understands a wide context of scenes. 5. Works very well with my cellulite lora, and allows alot of control on the bodyshape when adjusting strength on both loras individually. For the examples I used Krea2 Basemodel: fineporn_INT8_v4, CLIP: qwen-3-vl-4b-heretic_int8, VAE: wan_2.1_vae, inference steps: 8, sampler: euler_ancestral, scheduler: sgm_uniform, resolution: 1080x1920. -- Limitations: ❌❌❌ This lora model has a hard time with upside-down poses, or concepts like contortion, suspension and other concepts where the person is in a unnatural position. For this using a secondary lora trained on such images is needed. -- About this version v1 uses a combination of both photos and AI images. Mostly used on a wide variety of token style captions, added manually. Dataset re-iterated over 3 times, during a period of 6 months.