Moss-v2.0
推荐配置|Recommended configuration:
Sampler: Euler, Schedule type: MOSS (forge neo)
text_encoder:
qwen3.5_9b_int8_convrot.safetensors
moss_te_heretic_te_int8_convrot.safetensors
快速版:Steps: 8-12, CFG scale: 1, Size: 768x1280
8-step lora: MOSS_Image_SFT_8_steps_dmd_Latest.safetensors
高画质版:Steps: 30-50, CFG scale: 3-3.5, Size: 864x1536 | 1152x2048 (长宽保持64的倍数)
负面提示词推荐:blurry, low quality, jpeg artifacts, watermark, text, logo, deformed hands, extra fingers, fused fingers, bad anatomy, mutated limbs
融合菜单:
moss_image_sft_bf16.safetensors
<lora:RadianceChromeVoluptuous_MOSS-Image_base_v1.0:0.4>
<lora:moodyBetterNudes_v1:0.35>
<lora:v67-film-noir-moss-image-base-epoch-01:0.4>
<lora:mossimage_highlandscapeV1:0.4>
Moss-v1.0
推荐配置|Recommended configuration:
Sampler: Euler, Schedule type: MOSS (forge neo)
text_encoder: qwen3.5_9b_int8_convrot.safetensors
8-step lora: Moss_SFT_8_steps_dmd_v0.1
快速版:Steps: 8, CFG scale: 1, Size: 832x1216
高画质版:Steps: 30-50, CFG scale: 3, Size: 864x1536 | 1152x2048 (长宽保持64的倍数)
负面提示词推荐:blurry, low quality, jpeg artifacts, watermark, text, logo, deformed hands, extra fingers, fused fingers, bad anatomy, mutated limbs
融合菜单:
moss_image_sft_bf16.safetensors +
<lora:hina_MOSSImageSFT_asianMix_v2.0_lora:0.4> +<lora:RadianceChromeVoluptuous_MOSS-Image_base_v1.0:0.4> +
<lora:MysticXXX-Moss-V1:0.35> +
<lora:moodyBetterNudes_v1:0.35> +
<lora:v67-film-noir-moss-image-base-epoch-01:0.4> +<lora:mossimage_highlandscapeV1:0.4> +
<lora:moss-masterpieces-moss-timestep:0.25>
Krea2-v1.0
推荐配置|Recommended configuration:
Euler, Schedule type: Simple
step: 8-10, CFG: 1, 832x1216
vae: qwen_image_vae.safetensors
text_encoder:krea2_qwen3vl_4b_fp8_scaled.safetensors
Anima-v1.0
推荐配合:BlackStone HHP_v2(开发中)
推荐配置|Recommended configuration:
Euler a, Schedule type: Normal
step: 26, CFG: 4.5, Shift:3, 832x1216
vae: qwen_image_vae.safetensors
text_encoder:qwen_3_06b_base.safetensors
il_v5.0
推荐配置|Recommended configuration:
Euler a, Schedule type: Normal
step: 50, CFG: 2.5, 832x1216/680x1208 Distilled CFG Scale:3.5
Hires.fix, Upscale by: 2 hires steps: 16, Denoising strength: 0.3, Hires CFG: 2.5, Hires Distilled CFG Scale:3.5
vae: 已自带
il_v4.1
推荐配置|Recommended configuration:
Euler a, Schedule type: Normal
step: 36, CFG: 5, 832x1216/680x1208 Distilled CFG Scale:3.5
Hires.fix, Upscale by: 1.5 hires steps: 20, Denoising strength: 0.35, Hires CFG: 5, Hires Distilled CFG Scale:3.5
vae: 已自带
推荐配合BlackStone HHP-SET_v4并使用下面提示词:
prompt:
__HHP-SET/Global/randomEroticArt__
<lora:ILXL_Realism_Slider_v1 by klaabu @ civitai.com - thanks, cunt!:3.3> <lora:cknb02_stabilizer_v0.304a_fp16:0.35> <lora:DynamicPoseIL2att_alpha1.0_rank4_noxattn_900steps:1>
<lora:FLMFX:0.35> FLMFX, Film photo, Analog photography, 35mm film, Grainy texture, Film grain, Soft lighting, Vintage look, Washed colors, Light leaks, Natural shadows, <lora:Mirazhæl:0.35> miraz, scenery,
negative_prompt:
__HHP-SET/Global/Negative__
Description
FAQ
Comments (8)
I want to know how to generate faces with specific styles, because using terms like "Asian" or "European" doesn't work very well.
这个模型主要是偏亚洲人长相模型,可能因为这个原因导致“人种”这类关键词不敏感。
@ggyydream 我换了好几种提示词,无论是三角脸、鹅蛋脸、亚洲人和负面填西式都没有办法出那种“中式网红脸”(或者说是偏东方印象的,我不知道怎么说),基本上带点欧式的感觉,反正肯定不是中式,所以想问一下作者有没有什么这方面的提示词建议
test | Civitai
我发了一个我跑的图+提示词,但是那看起来很显然不是偏向亚洲面孔,有什么办法吗?谢谢
@zhuyun 我混合的模型中确实是有亚洲面孔的模型,但混合之后会导致面孔更像是混血的样子。你可以尝试一下这个LORA:https://civitai.com/models/2098342/realistic-asian-beauty-enhancer 看是否有所帮助。
The second test of this model - https://civitai.red/articles/31866/the-one-illustrious-sheckpoint-model-the-killer-of-all-sdxl-models-test-2
非常感谢你带来的专业比较测试,能得到你的赞誉,我很开心 :)
I did a test https://civitai.red/articles/31880/the-one-illustrious-checkpoint-the-absolute-flux-killer-is-finally-here-3 of CatPony-Real Illustrious model and a comparison with FLUX, especially for you.



















