Mei_Washio (鷲尾めい)

Imitation model. Modelled after Japanese AV J-cup 0ppai-goddess Mei Washio (鷲尾めい), previously a.k.a Jun Kakei (筧ジュン).

Back to creating big 0ppai LoRAs.

Easily one of the high ranked J-cup goddess: Angelic face, devilish figure. Mei debuted in the industry back in 2019, under the alias of Jun Kakei with powerhouse company S1. She changed over to 'Mei Washio' after swapping her entertainment company. Commonly contesting the title of J-cup goddess with Nagi Hikaru, Mei's filmography is equally impressive, having tried most of the popular genres. I personally see her body figure and vibe complementary to Nagi's strength: she's a sligthly thicc-er version of Nagi, bustier, and maybe with abit 'baby-face'.

Now what does a LoRA modelling both Nagi and Mei look like? Hmm..

~~~~~~~~~~~~~~~~~~~~~~~~~~
Recommended LoRA strength: 0.8 (0.6-1.0 are plausible range dependent on checkpoints; higher values >1 [unintensionally] increases picture brightness slightly..)

Main Prompt:

meiwashio_, , short hair, parted lips, bangs

variations: single hair bun, hair bun, tied hair, high ponytail, tied hair, brown hair, red lips, lipstick,

variations: parted lips, lipstick, makeup, blush, teeth, smile

Use of prompts depend on checkpoints, I didnt actually bother, since the main prompts works fine. (see tips below)

red lips will make her look slightly more mature. So check with your checkpoint on adjusting this back.

note that its "" not "" as always, since there's no doubt those are BIG on Mei.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Checkpoints that work well for this LoRA:

1) Realisian. Gravure-like photo quality. Best in Slot as usual.

2) RealisticMix. Consistent performance, sicne this checkpoint loves big 0ppai.

3) Ultimate- Realistic. Surprisingly well, considering its not originally intended for realistic-image generation (or does it? confused language)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Note for achieving good resemblance:

Very easy plug-and-play LoRA. You can choose to either use a custom ADetailer prompt, or just use the main prompt.

Very straightforward to use. If you are lazy, you can just use the main prompt to control for everything and apply Adetailer to fix the faces.

Else, you can separate the prompts into 2 stages:

1. Generate the main prompt using a LoRA strength of 0.8-1.0. so its probably safe to say that this weight is stable for use.

2. Now depending on what you use for (1), turn on Adetailer, and put something like

"face of meiwashio_, short hair, parted lips, bangs {use identical hair and makeup prompts to main prompt}<lora here>"

and set denoise strength to 0.4. (adjust accordingly from there)

It is still better to use AD at baseline generation (512,768 etc) and then upscale it later.

Recommended approach: is to not use HiRes, but upscale it with img2img instead.

Use a denoise strength of 0.20 -0.25 (anything higher will overcorrect face during the process), and turn off Adetailer. I prefer 0.2 since I use foolhardy-reMarci.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Issues:

No issues found as of yet.

Post-production edits/Reproducibility:

No edits. all img2img original upscales.

Happy prompting. And do share [SFW] creations~