A completely transparent white dress

My model, all free, open for public welfare, prohibited for commercial use, please consciously abide by local laws and regulations when using it. If you like it, remember to give it a thumbs up and a feedback, thank you for your support.

Additional notes:If you find that my lora model, both others and the author are using it well, you have added trigger words, and are using it correctly, but stillit doesn't work at all, it is likely because I am modifying the instructions for using lora, during the review, it will not take effect. Each modification will be reviewed in about half an hour to an hour, and it will return to normal after a while.(The title is correct, the cover is messed up, remember to use situation 4, that is the main body. Damn! While I was typing this paragraph, the cover got messed up again, with tears I replaced it with a low-quality bear image, tears in eyes, Buddha! Got messed up again, had to switch to a messed up image)

Reference weight: 0.7       Card drawing required     Adjustment range 0.6-0.8    

Core Trigger Words: (Choose different core trigger word combinations according to different needs, beginners are recommended to directly choose: Situation 4)

Situation 1:large breasts, white dress, see-through,(sitting or standing upper body change, remove high heels)

Situation 2:large breasts, white dress, see-through, high heels,(Use separately, sitting or standing full body image)

Situation 3:large breasts, cleavage, covered nipples, spaghetti strap, white dress, see-through, see-through silhouette, covered navel, high heels, realistic, (Start perspective view, turn into full transparent white spaghetti strap dress, with nsfw elements, recommend downloading and using at this point, some of my images surprisingly passed review, magical, remember to remove the "nsfw" word from negative prompt words when using it.)

Situation 4: (Highly recommended)large breasts, cleavage, covered nipples, spaghetti strap, white dress, see-through, see-through silhouette, covered navel, barefoot, showering, shower head, hand on own chest, water, wet clothes, wet dress, bathroom, (Bathroom bathing fully transparent wet body temptations, revealing all, full nsfw elements, recommend downloading and using at this point, good luck with the images passing review)

Situation 5:spread legs, kneeling, crossed legs, from behind,(Spread legs, kneeling, crossed legs, from behind, alternative actions and poses, remember, if you use a kneeling position, you must use the dimensions 768x1024, show the face in the image)

Recommended models:

Realistic: majicMIX realistic Orange Writing_v7 (My favorite general realistic bottom template, other models look fancy but after processing, compatibility issues arise, many bugs, not universal, of course, this is my personal feeling, no offense.)

Anime: AstrAnime 2.5D Anime(Presents excellent transparent wet effect), mix Ultimate Colorful Comic(Bright colorful visuals), White Castle's Thick Line 2.5D(Similar to thick line comic style), Girl YesMix v1.6(original)(Most universal, I often use it as an anime bottom template), TMND-Mix(After adding vae, it has a fresh and bright color), Chinese Wind 3 GuoFeng3 V3.4(The most universal 2.5d or 3d model, I often use it as a 3d Chinese animation bottom template) and so on, all have good performance. Also, some models may require the addition of a vae, if so, go to the author's page for instructions

Realistic, Anime, 2.5D are all acceptable

Note: Ifchanging the face, recommend 768x1024 size, using sitting or half-standing poses, use solo or full-body images, 512x1024 is the best choice.

When using HD repair, image texture and output rate will be significantly improved, recommend using the upscale algorithm R-ESRGAN_4x+, repair weight 0.33,if it is in anime style, choose the upscale algorithm corresponding to anime style R-ESRGAN_4x+ Anime6B, repair weight 0.33. For experienced users, consider using 0.35 or 0.75, but the effect is not guaranteed, based on current tests, I still recommend the repair weight of 0.33.

For many newcomers, let me explain: The examples are for reference only, but without adding the trigger words I provided, it is impossible to generate images. Also, I suggest simplifying the description of image quality, AI is like a person, the more descriptive words there are, the more attention is diverted, the easier it is to lose focus and the expression becomes mediocre. Just like me, provide key image quality description words for two to three sets is enough. Don't add a bunch of lora, add two or three compatible ones appropriately, that's enough, except for experienced users, many lora are incompatible with each other, which will distract AI's attention, dilute weight allocation, not to mention the collapse caused by incompatibility. Also, don't look for so-called powerful bottom models, they often come with various filters and special visual expressions, which makes lora unable to adapt, except for experienced users. For newcomers, when looking for a large model or bottom model, the top priority is compatibility, if you can't use it, it's useless, the more compatible, the more lora you can use. Unless you are using that model as a bottom model to specifically train lora for it. When generating images using lora, the top priority is still the models recommended by the author after testing, or the bottom models used to train lora, which means large models. Experienced users need not read this, it's for beginners who have just started AI drawing.

Example 1: (Realistic)

Front view:

1 girl, solo, perfect face, look at viewer, sitting, shower head, hand on own chest,
large breasts, cleavage, covered nipples, spaghetti strap, white dress, see-through, see-through silhouette, covered navel, barefoot, showering, water, wet clothes, wet dress, bathroom,
masterpiece, best quality, highly detailed,

Negative: Negative prompt: facess, (((censored))), (Multiple Girls), (bad hands), (bad feet), (bad legs), (bad penis), (Extra legs), (missing legs), (worst quality: 1.4), (low quality: 1.4), (normal quality: 1.4), low res,

Parameters: Steps: 20, Sampler: DDIM, CFG scale: 7.0, Seed: 2898232163, Size: 512x1024, Model hash: 7c819b6d13, Model: majicMIX realistic Orange Writing_v7.safetensors, Denoising strength: 0.33, Clip skip: 2, RNG: CPU, Lora 1: Fully transparent white skirt, Lora Hash 1: 6839e19c29, Lora Weight 1: 0.7, vae_name: automatic, Hires resize: 1024x2048, Hires steps: 20, Hires upscaler: R-ESRGAN_4x+

Example 2: (Anime)

Front view: 1 girl, solo, perfect face, look at viewer, sitting, shower head, hand on own chest,
large breasts, cleavage, covered nipples, spaghetti strap, white dress, see-through, see-through silhouette, covered navel, barefoot, showering, water, wet clothes, wet dress, bathroom,
masterpiece, best quality, highly detailed,

Negative: Negative prompt: facess, (((censored))), (Multiple Girls), (bad hands), (bad feet), (bad legs), (bad penis), (Extra legs), (missing legs), (worst quality: 1.4), (low quality: 1.4), (normal quality: 1.4), low res,

Parameters: Steps: 20, Sampler: Euler a, CFG scale: 7.0, Seed: 699774074, Size: 512x1024, Model hash: 5b53949fc6, Model: AstrAnime 2.5D Anime_V6.safetensors, Denoising strength: 0.33, Clip skip: 2, ENSD: 31337, RNG: CPU, Lora 1: Fully transparent white skirt, Lora Hash 1: 6839e19c29, Lora Weight 1: 0.7, vae_name: automatic, Hires resize: 1024x2048, Hires steps: 20, Hires upscaler: R-ESRGAN_4x+ Anime6B