使用介绍
关于2.0版本
对 NSFW 不太敏感。主要根据要求提供 NSFW 内容。该模型非常强大,所以我这样做是为了防止 NSFW 内容在不合适时干扰图像创建。总体改进,根据要求提供大量优质 NSFW 内容。
我不会说英语,所以我就长话短说吧。这是一个 NSFW SDXL 模型。基于 bigASP、NatViS:Natural Vision、LUSTIFY!以及我无处可寻的构建,感谢创建这些模型的人。
小心这些暗示,模特对 NSFW 非常敏感。她一直想这样做……这就是为什么她经常制造有趣的情况)))。
我不是这方面的专家,但我认为有些事情是有效果的,不要太苛刻地评判。
我还有其他模型,如果您喜欢这个模型,我会发布更多。
推荐设置:
Sampler:DPM++ 2M SDE/DPM++ 3M SDE,
Scheduler:Exponential/Karras,
Steps:25-55,
Cfg:2.5–6。我主要使用 3.5
Hires steps:10
Clip skip 2
Noise multiplier for img2img 1.1
ADetailer detector:mediapipe_face_full
Highres.fix:分辨率提升1.5,降噪~0.4。
对于远距离人物的照片,您可能需要使用 highres.fix 和/或 adetailer。
您可以尝试其他设置…
作品参照
Lionel Messi
Negative prompt: unrealistic, cartoon, 3d, 2d, Anime, CGI, badhandv4, By bad artist -neg, EasyNegative, EasyNegativeV2, FastNegativeV2, negative_hand-neg, ng_deepnegative_v1_75t (1), verybadimagenegative_v1.3, (worst quality:2), (low quality:2), (normal quality:2), (uneven eyes), lowres, normal quality, bad anatomy, bad face, (uneven eyes), paintings, ugly, bad hands, rolls, extra arms, extra eyes, extra anatomy, severed arm, bad arm, asymmetrical eyebrows, big mouth, deformed anatomy Unrealistic, low quality, simple, blurry, deformed nidfy parts, Low quality, blurry, extra fingers, badhandv4, By bad artist -neg, EasyNegative, EasyNegativeV2, FastNegativeV2, negative_hand-neg, ng_deepnegative_v1_75t (1), verybadimagenegative_v1.3, (worst quality:2), (low quality:2), (normal quality:2), (uneven eyes), lowres, normal quality, bad anatomy, bad face, (uneven eyes), paintings, ugly, bad hands
Steps: 55, CFG scale: 3.5, Sampler: DPM++ 2M SDE, Seed: 3547917873, VAE: sdxl_vae Clarity.safetensors, Size: 768x1152, Model: Bond 2.0 NSFW, Version: v1.10.1, FP8 weight: Enable for SDXL, Model hash: 40b9cce605, Hires steps: 10, Hires upscale: 1.5, Schedule type: Karras, Hires upscaler: OmniSR_X4_DIV2K, ADetailer model: mediapipe_face_full, ADetailer version: 24.6.0, Denoising strength: 0.4, ADetailer CFG scale: 5.0, ADetailer mask blur: 4, ADetailer model 2nd: mediapipe_face_mesh_eyes_only, ADetailer confidence: 0.75, Style Selector Style: base, ADetailer dilate erode: 4, Style Selector Enabled: True, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, ADetailer mask min ratio: 0.008, Style Selector Randomize: False, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, Cache FP16 weight for LoRA: True, ADetailer inpaint width 2nd: 768, ADetailer denoising strength: 0.6, ADetailer inpaint height 2nd: 768, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.5, ADetailer use separate CFG scale: True, ADetailer inpaint only masked 2nd: True, ADetailer use inpaint width height 2nd: True, Clip skip: 2
"High resolution image featuring a stunning fashion model dressed in an elegant ensemble inspired by the natural world and earth tones. The model's skin is highlighted with intricate details that capture her unique features and heritage. The attire consists of a dress that appears to be made from nature-inspired materials, ranging from crimson to deep red hues. The background is depicted as a reddish-toned backdrop with raw textures representing broken walls, creating an atmospheric effect. This composition enhances the model's beauty and showcases her versatility in fashion and style.".
Negative prompt: child, loli, low quality, face asymmetry, eyes asymmetry, deformed eyes, cgi, plastic, watermark,
Steps: 55, CFG scale: 3.5, Sampler: DPM++ 2M SDE, Seed: 4203253096, VAE: sdxl_vae NatViS.safetensors, Size: 768x1152, Model: Bond 2.0 NSFW, Version: v1.10.1, FP8 weight: Enable for SDXL, Model hash: 40b9cce605, Hires steps: 10, Hires upscale: 1.5, Schedule type: Karras, Hires upscaler: OmniSR_X4_DIV2K, ADetailer model: mediapipe_face_full, ADetailer version: 24.6.0, Denoising strength: 0.4, ADetailer CFG scale: 5.0, ADetailer mask blur: 4, ADetailer model 2nd: mediapipe_face_mesh_eyes_only, ADetailer confidence: 0.3, Style Selector Style: base, ADetailer dilate erode: 4, Style Selector Enabled: True, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, Style Selector Randomize: False, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, Cache FP16 weight for LoRA: True, ADetailer inpaint width 2nd: 768, ADetailer denoising strength: 0.6, ADetailer inpaint height 2nd: 768, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.5, ADetailer use separate CFG scale: True, ADetailer inpaint only masked 2nd: True, ADetailer use inpaint width height 2nd: True, Clip skip: 2
._Emotions_defiance512, Biomechanical charlize theron hacker - shaman of the sisterhood of cables wearing cable - dress made of cables with biomechanical filigree, super haute couture highly detailed eye, highly detailed lips, highly detailed skin, bokeh, lenseflare, by Emil Melmoth, Marcin Nagraba, Rebecca Millen, shot with Hasselblad X1D - 50c
Negative prompt: child, loli, low quality, face asymmetry, eyes asymmetry, deformed eyes, cgi, plastic, watermark
Steps: 55, CFG scale: 3.5, Sampler: DPM++ 2M SDE, Seed: 1548868102, VAE: sdxl_vae NatViS.safetensors, Size: 768x1152, Model: Bond 2.0 NSFW, Version: v1.10.1, FP8 weight: Enable for SDXL, Model hash: 40b9cce605, Schedule type: Karras, ADetailer model: mediapipe_face_full, ADetailer version: 24.6.0, ADetailer CFG scale: 5.0, ADetailer mask blur: 4, ADetailer model 2nd: mediapipe_face_mesh_eyes_only, ADetailer confidence: 0.3, Style Selector Style: base, ADetailer dilate erode: 4, Style Selector Enabled: True, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, Style Selector Randomize: False, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, Cache FP16 weight for LoRA: True, ADetailer inpaint width 2nd: 768, ADetailer denoising strength: 0.6, ADetailer inpaint height 2nd: 768, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.5, ADetailer use separate CFG scale: True, ADetailer inpaint only masked 2nd: True, ADetailer use inpaint width height 2nd: True, Clip skip: 2
使用介绍
我不会说英语,所以我就长话短说吧。这是一个 NSFW SDXL 模型。基于 bigASP、NatViS:Natural Vision、LUSTIFY!以及我无处可寻的构建,感谢创建这些模型的人。
小心这些暗示,模特对 NSFW 非常敏感。她一直想这样做……这就是为什么她经常制造有趣的情况)))。
我不是这方面的专家,但我认为有些事情是有效果的,不要太苛刻地评判。
我还有其他模型,如果您喜欢这个模型,我会发布更多。
推荐设置:
采样器:DPM++ 2M SDE/DPM++ 3M SDE,
调度器:Exponential/Karras,
步骤:25-55,
Cfg:2.5–6。我主要使用 3.5
Hires steps:10
Clip skip 2
Noise multiplier for img2img 1.1
ADetailer detector:mediapipe_face_full
Highres.fix:分辨率提升1.5,降噪~0.4。
对于远距离人物的照片,您可能需要使用 highres.fix 和/或 adetailer。
您可以尝试其他设置…
作品参照
Julianne Moore sits on a park bench, side by side with Daniel Craig James Bond
Negative prompt: (worst quality, low quality, normal quality, lowres, 2d, 3d), illustration, cartoons, painting, (laugh line), ugly face, old face, bad hands, bad anatomy, (monochrome, grayscale, sepia), strabismus, text, watermark, missing limb, sketch, double navel, collapsed eyeshadow, multiple eyebrows, freckles, NSFW, child, loli, cgi, worst quality, plastic, watermark, fat, face asymmetry, eyes asymmetry, deformed eyes
Steps: 55, CFG scale: 3.5, Sampler: DPM++ 2M SDE, Seed: 1312602505, VAE: sdxl_vae Clarity.safetensors, Size: 768x1152, Model: Bond_10, Version: v1.10.1, FP8 weight: Enable for SDXL, Model hash: 65decc337c, Hires steps: 10, Hires upscale: 1.5, Schedule type: Karras, Hires upscaler: OmniSR_X4_DIV2K, ADetailer model: mediapipe_face_full, ADetailer version: 24.6.0, Denoising strength: 0.4, ADetailer CFG scale: 5.0, ADetailer mask blur: 4, ADetailer model 2nd: mediapipe_face_mesh_eyes_only, Hires schedule type: Karras, ADetailer confidence: 0.75, Style Selector Style: base, ADetailer dilate erode: 4, Style Selector Enabled: True, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, ADetailer mask min ratio: 0.008, Style Selector Randomize: False, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, Cache FP16 weight for LoRA: True, ADetailer inpaint width 2nd: 768, ADetailer denoising strength: 0.4, ADetailer inpaint height 2nd: 768, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.5, ADetailer use separate CFG scale: True, ADetailer inpaint only masked 2nd: True, ADetailer use inpaint width height 2nd: True
Cinematic-style photograph, digital medium. Close-up of a woman partially submerged in water with intense, expressive eyes peering above the surface. Streamlined dark hair slicked back, reflecting natural light. Surrounding environment features lily pads and blurred aquatic background with rich, earthy greens and muted browns. Dramatic play of light and shadow on water and woman's face creating a mysterious ambiance. High contrast, meticulous detail, capturing an enigmatic, suspenseful moment.
Negative prompt: child, loli, cgi, worst quality, plastic, watermark, fat, face asymmetry, eyes asymmetry, deformed eyes
(worst quality, low quality, normal quality, lowres, 2d, 3d), illustration, cartoons, painting, (laugh line), ugly face, old face, bad hands, bad anatomy, (monochrome, grayscale, sepia), strabismus, text, watermark, missing limb, sketch, double navel, collapsed eyeshadow, multiple eyebrows, freckles,
Steps: 55, CFG scale: 3.5, Sampler: DPM++ 2M SDE, Seed: 543090162, VAE: sdxl_vae NatViS.safetensors, Size: 768x1152, Model: bond10NSFW_noVAE, Version: v1.10.1, FP8 weight: Enable for SDXL, Model hash: 0bee2d7082, Hires steps: 10, Hires upscale: 1.5, Schedule type: Karras, Hires upscaler: OmniSR_X4_DIV2K, ADetailer model: mediapipe_face_full, ADetailer version: 24.6.0, Denoising strength: 0.4, ADetailer CFG scale: 5.0, ADetailer mask blur: 4, ADetailer model 2nd: mediapipe_face_mesh_eyes_only, Hires schedule type: Karras, ADetailer confidence: 0.75, Style Selector Style: base, ADetailer dilate erode: 4, Style Selector Enabled: True, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, ADetailer mask min ratio: 0.005, Style Selector Randomize: False, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, Cache FP16 weight for LoRA: True, ADetailer inpaint width 2nd: 768, ADetailer denoising strength: 0.6, ADetailer inpaint height 2nd: 768, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.5, ADetailer use separate CFG scale: True, ADetailer inpaint only masked 2nd: True, ADetailer use inpaint width height 2nd: True, Clip skip: 2
Close-up shot of Rafael Nadal's determined face as he grips his tennis racket tightly, sweat glistening on his forehead under the bright sun. His eyes are fixed intensely on the ball, his jaw set in a fierce concentration. The soft focus background blurs the lush green tennis court, emphasizing the athlete's intense focus.
Negative prompt: child, loli, cgi, worst quality, plastic, watermark, fat, face asymmetry, eyes asymmetry, deformed eyes,
(worst quality, low quality:1.4),(Grayscale, Monochrome:1.1),cropped,low color,text,(NSFW:1.3),(worst quality:2),(low quality:2),(normal quality:2),(grayscale),acne,age spots,(Ugy:1.331),(copy:1.331),(morbid:1.21),(deformity:1.21),(degeneration:1.331),variant hand,(poorly drawn:1.5),(poor anatomy:1.21),(poor ratio:1.331),additional limbs,(disfigurement:1.331),(missing arm:1.331),(multiple legs:1.331),(combined finger:1.61051),(multiple fingers:1.61051),(blurred vision:1.331),low hand,bad hand,less finger,more finger,bad hand,less finger,(multiple arms and legs),,nsfw
Steps: 50, CFG scale: 3.5, Sampler: DPM++ 2M SDE, Seed: 3415715549, VAE: sdxl_vae NatViS.safetensors, Size: 768x1152, Model: bond10NSFW_noVAE, Version: v1.10.1, FP8 weight: Enable for SDXL, Model hash: 0bee2d7082, Hires steps: 10, Hires upscale: 1.5, Schedule type: Karras, Hires upscaler: OmniSR_X4_DIV2K, ADetailer model: mediapipe_face_full, ADetailer version: 24.6.0, Denoising strength: 0.4, ADetailer CFG scale: 5.0, ADetailer mask blur: 4, ADetailer model 2nd: mediapipe_face_mesh_eyes_only, Hires schedule type: Karras, ADetailer confidence: 0.75, Style Selector Style: base, ADetailer dilate erode: 4, Style Selector Enabled: True, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, ADetailer mask min ratio: 0.005, Style Selector Randomize: False, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, Cache FP16 weight for LoRA: True, ADetailer inpaint width 2nd: 768, ADetailer denoising strength: 0.6, ADetailer inpaint height 2nd: 768, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.5, ADetailer use separate CFG scale: True, ADetailer inpaint only masked 2nd: True, ADetailer use inpaint width height 2nd: True, Clip skip: 2